View U.S. Patent No. 10124008 in PDF
I1111111111111111 1111111111 111111111111111 IIIII 11111 111111111111111 IIII IIII USO 10124008B2 c12) United States Patent (IO) Patent No.: US 10,124,008 B2 Lakkaraju et al. (45) Date of Patent: Nov. 13, 2018
(54) USE OF INHIBITORS OF ACID (56) References Cited SPHINGOMYELINASE TO TREAT ACQUIRED AND INHERITED RETINAL FOREIGN PATENT DOCUMENTS DEGENERATIONS WO 9718815 A2 5/1997 (71) Applicant: Wisconsin Alumni Research WO 0007584 A2 2/2000 Foundation, Madison, WI (US) WO 2005097121 Al 10/2005 WO 2009023299 A2 2/2009 (72) Inventors: Aparna Lakkaraju, Madison, WI (US); Kimberly A. Toops, Sun Prairie, OTHER PUBLICATIONS WI (US); Li Xuan Tan, Madison, WI (US) Van Leeuwen et al. Ophthalmology, 2004, vol. 111, pp. 1169-1175.* Kolzer et al. FEBS Letters, 2004, Iss. 559, pp. 96-98.* (73) Assignee: Wisconsin Alumni Research Opreanu et al. Diabetes, 2011, vol. 60, pp. 2370-2378.* Foundation, Madison, WI (US) Ambati, et al., Mechanisms of age-related macular degeneration, Neuron, 2012, 75(1), 26-39. ( *) Notice: Subject to any disclaimer, the term ofthis Arroyo, A 76-Year-Old Man With Macular Degeneration, JAMA, patent is extended or adjusted under 35 2006, 295(20):2394-2406. Barmada, et al., Autophagy induction enhances TDP43 turnover and U.S.C. 154(b) by O days. survival in neuronal ALS models, Nature Chemical Biology, 2014, (21) Appl. No.: 14/746,221 10, 677-685. Bentley, et al., Tarnoxifen Retinopathy: A Rare But Serious Com (22) Filed: Jun. 22, 2015 plication, British Medical Journal, 1992, 304:495-496. Cunningham, et al., Clofazamine-Induced Generalized Retinal Degen (65) Prior Publication Data eration, Retina, 1990, 10:131-134. Doukas, et al., Topical administration of a multi-targeted kinase US 2015/0366876 Al Dec. 24, 2015 inhibitor supresses choroidal neovascularization and retinal edema, Journal of Cellular Physiology, 2008, 216(1), 29-37. Edward, et al., Amelioration of Light-Induced Retinal Degeneraiton Related U.S. Application Data by a Calcium Overload Blocker, Arch. Ophthalmol., 1991, 109:554- 562. (60) Provisional application No. 62/015,753, filed on Jun. Ewe, et al., Bilateral Maculopathy Associated with Sertraline, 23, 2014. Australasian Psychiatry, 2014, 22(6):573-575. Gannnons, Topical antiangiogenic SRPKl inhibitors reduce choroidal (51) Int. Cl. neovascularization in rodent models of exudative AMD, IOVS, A61K 31155 (2006.01) 2013, 54(9), 6052-6062. A61K 31/136 (2006.01) Grumati, et al., Autophagy is defective in collagen VI muscular A61K 9/00 (2006.01) dystrophies, and its reactivation rescues myofiber degeneration, A61K 311454 (2006.01) Nature Medicine, 2010, 16, 1313-1320. A61K 311445 (2006.01) Gulbins, et al., Acid sphingomyelinase-ceramide system mediates A61K 45106 (2006.01) effects of antidepressant drugs, Nature Medicine, 2013, 19, 934- A61K 31/135 (2006.01) 938. A61K 311137 (2006.01) Hardisty, et al., Citaloprarn-Associated Central Retinal Vein Occlu A61K 31/138 (2006.01) sion, Int. Ophthalmol., 2009, 29:303-304. Jimenez-Sanchez, The Hedgehog signaling pathway regulates A61K 31/343 (2006.01) autophagy, Nature Communications, 2012, 3(1200), 1-11. A61K 3114422 (2006.01) Kirkegaard, et al., Hsp70 stabilizes lysosomes and reverts Niemann A61K 311451 (2006.01) Pick disease-associated lysosomal pathology, Nature, 2010, 463(7280), A61K 3114525 (2006.01) 549-553. A61K 31146 (2006.01) Kolzer, et al., Interactions of acid sphingomyelinase and lipid A61K 3115415 (2006.01) bilayers in the presence of the tricyclic antidepressant desiprarnine, A61K 311675 (2006.01) FEBS Letters, 2004, 559, 96-98. (52) U.S. Cl. (Continued) CPC ...... A61K 31155 (2013.01); A61K 9/0048 (2013.01); A61K 31/135 (2013.01); A61K Primary Examiner - Samira J Jean-Louis 31/136 (2013.01); A61K 311137 (2013.01); (74) Attorney, Agent, or Firm -Quarles & Brady LLP A61K 31/138 (2013.01); A61K 31/343 (2013.01); A61K 311445 (2013.01); A61K (57) ABSTRACT 3114422 (2013.01); A61K 311451 (2013.01); A method of treating retinal diseases is disclosed that A61K 311454 (2013.01); A61K 3114525 includes the step of administering an effective amount of a (2013.01); A61K 31146 (2013.01); A61K composition including an ASMase inhibitor to a retinal 3115415 (2013.01); A61K 311675 (2013.01); disease patient, wherein at least one disease symptom is A61K 45106 (2013.01) either lessened or progression of the symptom is delayed. (58) Field of Classification Search CPC ...... A61K 31/55; A61K 31/136 10 Claims, 13 Drawing Sheets See application file for complete search history. (13 of 13 Drawing Sheet(s) Filed in Color) US 10,124,008 B2 Page 2
(56) References Cited Dannhausen, K., M. Karlstetter, A. Caramoy, C. Volz, H. Jagle, G. Liebisch, 0. Utermohlen, and T. Langmann. 2015. Acid sphingomyelinase (aSMase) deficiency leads to abnormal microglia OTHER PUBLICATIONS behavior and disturbed retinal function. Biochem Biophys Res Commun. 464:434-440. Kornhuber, et al., Functional inhibitors of acid sphingomyelinase Fan, J., B.X. Wu, and C.E. Crosson. 2016. Suppression of Acid (FIASMAs): A novel pharmacological group of drugs with broad Sphingomyelinase Protects the Retina from Ischemic Injury. Invest clinical applications, Cellular Physiology and Biochemistry, 2010, Ophthalmol Vis Sci. 57:4476-4484. 26, 9-20. Feskanich, D., E. Cho, D.A. Schaumberg, G.A. Colditz, and S.E. Lajunen, et al., Topical drug delivery to retinal pigment epithelium Hankinson. 2008. Menopausal and reproductive factors and risk of with microfluidizer produced small liposomes, European Journal of age-related macular degeneration. Arch Ophthalmol. 126:519-524. Pharmaceutical Sciences, 2014, 62, 23-32. Fox, T.E., X. Han, S. Kelly, A.H. Merrill, 2nd, R.E. Martin, R.E. Lakkaraju, et al., Low-density lipoprotein receptor-related protein Anderson, T.W. Gardner, and M. Kester. 2006. Diabetes alters mediates the endocytosis of anionic liposomes in neurons, Journal sphingolipid metabolism in the retina: a potential mechanism of cell of Biological Chemistry, 2002, 277(17), 15085-15092. death in diabetic retinopathy. Diabetes. 55:3573-3580. Lakkaraju, et al., The lipofuscin fluorophore A2E perturbs choles Fraser-Bell, S., J. Wu, R. Klein, S.P. Azen, and R. Varma. 2006. terol metabolism in retinal pigment epithelial cells, PNAS, 2007, Smoking, alcohol intake, estrogen use, and age-related macular 104(26), 11026-11031. degeneration in Latinos: the Los Angeles Latino Eye Study. Am J Le Guezennec, et al., Wipl-dependent regulation of autophagy, Ophthalmol. 141:79-87. obesity, and atherosclerosis, Cell Metabolism, 2012, 16(1), 68-80. Frouws, M.A., B.G. Sibinga Mulder, E. Bastiaannet, M.M. Zanders, Lee, et al., Acid sphingomyelinase modulates the autophagic pro M.P. van Herk-Sukel, E.M. de Leede, B.A. Bonsing, J.S. Mieog, cess by controlling lysosomal biogenesis in Alzheimer's disease, C.J. Van de Velde, and G.J. Liefers. 2017. No association between JEM, 2014, 211(8), 1551-1570. metformin use and survival in patients with pancreatic cancer: An Nixon, The role of autophagy in neurodegenerative disease, Nature observational cohort study. Medicine (Baltimore). 96:e6229. Medicine, 2013, 19, 983-997. Gandini, S., M. Puntoni, B.M. Heckman-Stoddard, B.K. Dunn, L. Pampliega, et al., Functional interaction between autophagy and Ford, A. DeCensi, and E. Szabo. 2014. Metformin and cancer risk ciliogenesis, Nature, 2013, 502, 194-200. and mortality: a systematic review and meta-analysis taking into Penfold, et al., Modulation of major histocompatibility complex account biases and confounders. Cancer Prev Res (Phila). 7:867- class II expression in retinas with age-related macular degeneration, 885. Invest Ophthalmol Vis Sci, 1997, 38(10), 2125-2133. Klein, R., B.E. Klein, S.C. Jensen, K.J. Cruickshanks, K.E. Lee, Petersen, et al., Transformation-associated changes in sphingolipid L.G. Danforth, and S.C. Tomany. 2001. Medication use and the metabolism sensitize cells to lysosomal cell death induced by 5-year incidence of early age-related maculopathy: the Beaver Dam inhibitors of acid sphingomyelinase, Cancer Cell, 2013, 24(3), Eye Study. Arch Ophthalmol. 119: 1354-1359. 379-393. Kornhuber, J., P. Tripal, M. Reichel, C. Muhle, C. Rhein, M. Radu, et al, Bisretinoid-mediated complement activation on retinal Muehlbacher, T.W. Groemer, and E. Gulbins. 2010. Functional pigment epithelial cells is dependent on complement factor H Inhibitors of Acid Sphingomyelinase (FIASMAs ): a novel pharma haplotype, Journal of Biological Chemistry, 2014, 289(13), 9113- cological group of drugs with broad clinical applications. Cellular 9120. physiology and biochemistry : international journal of experimental Roth, et al., Potent and selective inhibitors of acid sphingomyelinase cellular physiology, biochemistry, and pharmacology. 26:9-20. by bisphosphonates, Angewandte Chemie International Edition, Pikuleva, I.A., and C.A. Curcio. 2014. Cholesterol in the retina: The 2009, 48, 7560-7563. best is yet to come. Prog Retin Eye Res. Rotstein, et al., Regulating Survival and Development in the Retina: Suissa, S. 2012. Randomized Trials Built on Sand: Examples from Key Roles for Simple Sphingolipids, Journal of Lipid Research, COPD, Hormone Therapy, and Cancer. Rambam Maimonides Med 2010, 51:1247-1262. J. 3:e0014. Sander, 587: The Astemizole Retinopathy Trial: Effect of Astemizol Suissa, S., and L. Azoulay. 2012. Metformin and the risk of cancer: on Diabetic Macular, Investigative Ophthalmology & Visual Sci time-related biases in observational studies. Diabetes Care. 35 :2665- ence: Annual Meeting of the Association for Research in Vision and 2673. Ophthalmology, Association for Research in Vision and Ophthal Toops, K.A., L.X. Tan, Z. Jiang, R. Radu, and A. Lakkaraju. 2015. mology, 2000, 41(4):Sll4. Toops, et al., A detailed three-step protocol for live imaging of Cholesterol-mediated activation of acid sphingomyelinase disrupts intracellular traffic in polarized primary porcine RPE monolayers, autophagy in the retinal pigment epithelium. Mo! Biol Cell. 26: 1-14. Experimental Eye Research, 2014, 124, 74-85. Wu, B.X., J. Fan, N.P. Boyer, Rim. Jenkins, Y. Koutalos, Y.A. Toops, et al., Cholesterol-mediated activation of acid sphingomyelinase Hannun, and C.E. Crosson. 2015. Lack of Acid Sphingomyelinase disrupts autophagy in the retinal pigment epithelium, Molecular Induces Age-Related Retinal Degeneration. PLoS One. 10:e0 133032. Biology of the Cell, 2015, 26(1), 1-14. Amsterdam, J.D., D.J. Brunswick, L. Potter, A. Winokur, and K. Travis, et al. Diseases caused by defects in the visual cycle: Rickels. 1985. Desipramine and 2-hydroxydesipramine plasma lev Retinoids as potential therapeutic agents, Annu. Rev. Pharmacol els in endogenous depressed patients. Lack of correlation with Toxicol, 2007, 47, 469-512. therapeutic response. Archives of general psychiatry. 42:361-364. Tune, Maculopathy Following Extended Usage of Clomiphene Baumann, P., S. Ulrich, G. Eckermann, M. Gerlach, H.J. Kuss, G. Citrate, Eye, 2014, 28: 1144-1146. Laux, B. Muller-Oerlinghausen, M.L. Rao, P. Riederer, G. Zernig, Wikipedia, The Free Encyclopedia, "FIASMA", Current Revision, C. Hiernke, and g. Arbeitsgemeinschaft fur N europsychopharmakologie Edited by Klbrain (talk/ contribs), Aug. 2015. und Pharmakopsyhiatrie-Therapeutic Drug Monitoring. 2005. The Zhao, et al., mTOR-mediated dedifferentiation of the retinal pig AGNP-TDM Expert Group Consensus Guidelines: focus on thera ment epithelium initiates photoreceptor degeneration in mice, J Clin peutic monitoring of antidepressants. Dialogues in clinical neuro Invest., 2011, 121(1), 369-383. science. 7 :231-247. Zhou, et al., Complement activation by bisretinoid constituents of Tan, L.X., K.A. Toops, and A. Lakkaraju. 2016. Protective responses RPE lipofuscin, Invest Ophthalmol Vis Sci, 2009, 50(3), 1392-1399. to sublytic complement in the retinal pigment epithelium. Proceed PCT International Search Report and Written Opinion, PCT/US2015/ ings of the National Academy of Sciences of the United States of 036982, dated Sep. 24, 2015. America. 113:8789-8794. Busik, J.V., W.J. Esselman, and G.E. Reid.2012. Examining the role Toops, K.A., L.X. Tan, Z. Jiang, R.A. Radu, and A. Lakkaraju. of lipid mediators in diabetic retinopathy. Clin Lipidol. 7:661-675. 2015. Cholesterol-mediated activation of acid sphingomyelinase US 10,124,008 B2 Page 3
(56) References Cited
OTHER PUBLICATIONS disrupts autophagy in the retinal pigment epithelium. Molecular biology of the cell. 26:1-14.
* cited by examiner U.S. Patent Nov. 13, 2018 Sheet 1 of 13 US 10,124,008 B2
C 0 [J:l Co,11:r<>i -A2E
E
...... C.,~1:r~tQ f ••••••••• • ••••••••••AZE ...... T<)fiM, - + 4, + Bai' LC38--\ i.C:l-13--1:
G
2/l C] M,:J trf:alr::wnt D -r~.1~·!:~s
11111111111 <\)!"1'1$ + 13'.lf
Gor.tr~I fa2E
FIG. lA-lG U.S. Patent Nov. 13, 2018 Sheet 2 of 13 US 10,124,008 B2
FIG. 2 U.S. Patent Nov. 13, 2018 Sheet 3 of 13 US 10,124,008 B2
A B
I •
0 ~~ F.. s 'f<:
E
[illC,:mtml .A2E
?J&x ~ It iii~•
',~LJ... . . 1111. .1.~J ~~~&/4~~~P/4P*~$/4~t#~~~~0~~~~~0~~~~~~01~~~~0 <::,<;.';- f~--,;.~ tj·,-:§i""':..)·.,-::;::i~r::,'·<:J~ '."';.}\0 '1V·..::};> '),•~'? -=:::~r- ·-::;•<}S"\'4:-~~-.... ·{)· ~-{y '·<.:/'{,:;· {,>·\":,· -,~/ ,7~,·... ()·:.'t,•''t:,·~· {)·:-'."v. ❖ ·;;;,·;..·~~\)·Y\-.).~<)"1°''"
FIG. 3A-3E U.S. Patent Nov. 13, 2018 Sheet 4 of 13 US 10,124,008 B2
A i"too-,, P,:;;i,.:ir:i:u,J pr,(tm::-i RP;; B CD RPE fr<:,,Y: w;1rJ-t?,pr, nli<:<:i * - f;'.PE imr,, ABCA ,f/- :niN, _1!_:t
.Ait.ut~ A2E=
C ______c:wiP>L.______MJ.;______- c-:.mm;•1__ ___ f2;;. __ _
E zr -* l ·-r l:.~i. •:•:•:•: r•"-•••"-••···) :-
em1-ir-~i ..~,2e: ~~xi::!.<:!? ,.;;·~r Cr.:-rt:--t>i: 1'2t: •t.rx' f.k;f · · + ,. ·· ;- . 4 -· .. Bat--·•--+ - ❖ - + f G Cl.: jo.J ··--:;--- ii io.:z:g._. ii µ;:=::cu O.C FIG. 4A-4G U.S. Patent Nov. 13, 2018 Sheet 5 of 13 US 10,124,008 B2 (~.ri:rcr..:~i: K)w- C PJwlbidin Crmtmi Ctlr D At>tub ZO~t f.MP! Control AlE U18666A FIG. 5A-5D U.S. Patent Nov. 13, 2018 Sheet 6 of 13 US 10,124,008 B2 A ~T~iwt:.it.;-i.;:\i::: -AC-·-r$;b:;_;!:r,: ,:i 137 ""- (J'' Chrt1t::<: Gt}ilt.:'{;t ir,•,-. ~·,·· C D C •,' FlG. 6A-6G U.S. Patent Nov. 13, 2018 Sheet 7 of 13 US 10,124,008 B2 .A B FIG. 7A-7D U.S. Patent Nov. 13, 2018 Sheet 8 of 13 US 10,124,008 B2 A B D .. 5. -iO 1~ 20 $ i:) 15 j:; 'fol«! b,,,,if .1:~C;f:mt,i•,i. (IH11i T-:.'t;1t tff;f;k t~i:=::fJ!:fg"i;;~: 1:!f; a~. (~~:t:) FIG. 8A-8E U.S. Patent Nov. 13, 2018 Sheet 9 of 13 US 10,124,008 B2 _...... ·+ n~~iFr:arninQ { 0~$} .....- A ...... • o.2!;_...... Ggn~r;jt. .•.-.A~ Toiin'1> ❖ + + + ❖ + ❖ !:.l«f .,. + .,. + t...c;:;:s.-.:: lC:-m.:~ nc----===:;- C.~:1tn)l ;\2:f.. "'="""'=c------;;:"'.'\.;:'t, ~ + + + ·-r~Jfir"1:S -~- + .,. .. + rliif + • + {1 /4.... ffli S*i":ff.:l ~:f'i~8.,-jfJ ~fuhC,J:1 [~~ A2E AJE E :c';d·,; Hartl 1~·,·:.. 1·: i},:::..i -·-· -·- '... ·- G;.:«J f.x,~ lQS. .t{f£ -td~~ FIG. 9A-9E U.S. Patent Nov. 13, 2018 Sheet 10 of 13 US 10,124,008 B2 .A. MECHAM.SM OF AS MASE B. AUTOPHAG'flN C. f4.J70PHAGt lN .M.';Tf\lATfON BY Bf.SRETl'NOIDS NORMAL. Rf>£ R.P.E wm-t Bff ;;,;:-i;1{;,f ~.G? r«ix,mf1? ' :@; ~ 110!:·• 1 ::I:: 11:::::: _ I ;,i:,ei;tmtwi', etabl ,1 0j.':'l1$ffi!~ , m/croM,uJ&.:.; m1troi.1,t,ui1i,·, ('Y I .•·;;,;;·,,:«·»'"""l '"' ·•: ::;::~ FIG. l0A-l0C U.S. Patent Nov. 13, 2018 Sheet 11 of 13 US 10,124,008 B2 A O>+DAPI ClXWDAPI Abt"1'1"'· R~•E. C 0 8 D a:: - Reactive oxygen species Q ~ 0.6 C ~ ~ 0.4 0 :;; ti::: S 0.2· ~ OJ (.) 0.0 +NHS E 1 MHC-I1 expression ** 0. Control Des A2E A2E + Des FIGS. llA-llE U.S. Patent Nov. 13, 2018 Sheet 12 of 13 US 10,124,008 B2 A Predicted MW ~70 kDa / Observed MW -6-5 kDa Additional band -sa kDa j Actin B C 150 (1) ~ 125 FIGS. 12A-12C U.S. Patent Nov. 13, 2018 Sheet 13 of 13 US 10,124,008 B2 A B 50 0 fi5 0.30 s... a.. LL ~ 0.25 a.. LL O 0,20 w A2E +Des +Ast +Fex +Ser FIG. 13A-13C US 10,124,008 B2 1 2 USE OF INHIBITORS OF ACID including the daily phagocytosis and degradation of shed SPHINGOMYELINASE TO TREAT photoreceptor outer segments (Bok, 1993). The RPE is also ACQUIRED AND INHERITED RETINAL the primary site of damage in many retinal degenerative DEGENERATIONS diseases including AMD (Ambati and Fowler, 2012). The 5 RPE is a post-mitotic tissue and with age, undigested outer CROSS-REFERENCE TO RELATED segment components accumulate in the form oflipofuscin in APPLICATIONS RPE lysosomes. This lipofuscin is composed primarily of This application claims priority to U.S. Patent Ser. No. vitamin A metabolites called bisretinoids. A major compo 62/015,753, filed Jun. 23, 2014, which is incorporated by nent ofRPE lipofuscin is A2E, which is a Schiff-base adduct 10 reference herein. of vitamin A and ethanolamine. Although increasing levels of lipofuscin and A2E are hallmarks of retinal degenerations STATEMENT REGARDING FEDERALLY including AMD, Stargardt and Best diseases (Rattner and SPONSORED RESEARCH OR DEVELOPMENT Nathans, Nature Neuroscience, 2006 and Ambati and Fowler, supra), precisely how they harm the RPE and This invention was not made with government support. 15 promote vision loss is unclear. BACKGROUND Our data has shown that lipofuscin and A2E increase lysosomal cholesterol in a primary RPE cell-based model Over 30 million people currently suffer from AMD, the that recapitulates key features of AMD (Lakkaraju et al., most common cause of permanent vision loss among older 20 2007) and in a mouse model of Stargardt disease (Toops et adults. There are currently no approved therapies for geo al., 2015). Cholesterol accumulation caused by lipofuscin graphic atrophy, which affects 90% of AMD patients (Me and A2E interferes with multiple steps of autophagy, an leth et al., 2011), or for juvenile-onset macular degenera essential mechanism for clearing debris in the post-mitotic tions like Stargardt disease, which has a prevalence of 1 in RPE. High-speed live imaging showed that autophagosome 8000 children (Haji Abdollahi and Hirose, 2013). 25 biogenesis and trafficking is significantly disrupted in RPE This lack of effective treatments is largely due to limited insight into disease mechanisms. Current clinical trials for with the bisretinoid A2E due to disrupted transport of AMD and Stargardt disease are focused on stem cell trans autophagosomes. plants or intravitreal injections of antibodies to complement In the present study, our data demonstrate a mechanism by pathway proteins (clinicaltrials.gov), all of which require which A2E-induced excess lysosomal cholesterol activates periodic (monthly to quarterly) injections into the vitreous. 30 acid sphingomyelinase (AS Mase), the enzyme that hydro This mode of administration has been widely used to admin lyzes sphingomyelin to ceramide. Ceramide promotes acety ister VEGF antibodies to treat wet AMD. Intravitreal injec lation of tubulin on stabilized microtubules, leading to tions have numerous unacceptable consequences including impaired trafficking of autophagosomes. We show that high endophthalmitis, retinal detachment and increased risk of ceramide levels are directly responsible for the autophagic developing geographic atrophy (Lois, et al., 2013). 35 block in RPE with A2E because treatment with the AS Mase "Wet" AMD, also known as choroidal neovascularization, is caused by abnormal growth of choroidal blood vessels inhibitor desipramine (Kaizer et al., 2004), a tricyclic anti into the retina. Antibodies to vascular endothelial growth depressant, decreased tubulin acetylation, corrected traffick factor (VEGF) are currently used to prevent blood vessel ing defects and restored autophagic flux in the RPE. More- growth. The more chronic form of the disease is called "dry" 40 over, we show that desipramine is very effective in reversing AMD, or geographic atrophy, where progressive degenera pro-inflammatory conditions in RPE with bisretinoids by (i) tion of the retinal pigment epithelium and subsequent loss of increasing CD59 delivery to the plasma membrane; (ii) overlying photoreceptors leads to a slow decline in central, increasing the efficiency of membrane repair after comple high-resolution vision. There are no approved therapies for ment attack; (iii) limiting the production ofreactive oxygen dry AMD. 45 species; and (iv) decreasing MHC-II protein levels. A large-scale multicenter trial on high dose antioxidant and carotenoid supplements recently showed that the supple SUMMARY OF THE INVENTION ments slow the progression ofAMD from early to late stages but do not prevent disease onset. EMIXUSTAT, a visual According to a first aspect, a method of treating retinal cycle modulator that decreases the formation of A2E and 50 diseases includes the step of administering an effective lipofuscin is currently in Phase II trials. In Phase I trials, amount of a composition comprising an AS Mase inhibitor to EMIXUSTAT caused a dose-dependent delay in dark adap a retinal disease patient. At least one of retinal disease tation and dyschromatopsia (Kubota et al., 2012), suggesting symptom is lessened or progression of the symptom is that this could potentially limit its use in chronic diseases delayed. like dry AMD. 55 According to a second aspect, a method of treating Thus, there is a strong need for small molecule lipophilic macular degeneration in a subject in need thereof includes drugs with good safety profiles that can be administered the steps of identifying a subject with macular degeneration orally or as topical formulations to the eye to treat macular and administering to the subject a therapeutically effective dystrophies. Unlike hydrophilic drugs, lipophilic drugs can amount of a composition including an inhibitor of acid reach the retina by easily crossing the outer blood-retinal 60 barrier, which is formed by the tight junctions of the retinal sphingomyelinase activity. pigment epithelium (RPE). According to a third aspect, a method of treating macular degeneration in a subject in need thereof, includes the steps Retinal Disease Biology of identifying a subject with at least one of a) an accumu- 65 lation of soft, large drusen, b) increased fundus autofluo The RPE, which sits beneath the photoreceptors in the rescence, and c) delayed dark adaptation, and administering retina, performs numerous functions critical for vision, to the subject a therapeutically effective amount of a topical US 10,124,008 B2 3 4 composition comprising desipramine a composition com cantly lesser than corresponding condition without prising an inhibitor of acid sphingomyelinase activity. TO901317 (in red hatched bars), p<0.05. (F) Stills from live imaging of mRFP-GFP-LC3 in serum-starved RPE treated BRIEF DESCRIPTION OF THE DRAWINGS as indicated. (G) Quantification of EGFP (green) to mRFP 5 (red) fluorescence in (F). Mean±S.E.M., *, significantly This patent application file contains at least one drawing greater than all other treatments, p<0.05. executed in color. Copies of this patent or patent application FIG. SA-D. (A) Filipin staining to detect cholesterol pools publication with color drawings will be provided by the in cells treated with A2E or A2E and the LXRa agonist Office upon request and payment of the necessary fee. TO901317. Paraformaldehyde-fixed cells were incubated FIG. lA-G. Regulation of autophagy in the RPE (A) 10 with 50 µg/ml filipin (Sigma) for 45 min at room tempera Representative immunoblot and quantification of LC3B-II ture {Lakkaraju, 2007 #5829}. (B) Expression of beta levels in the RPE of 3- and 6-month-old wild-type (grey tubulin in primary RPE. Lysates from control or A2E-treated 1 bars) and ABCA4_ _ (black bars) mice. Significantly less (15 nM or 50 nM A2E for three weeks or 20 µM A2E for 6 than age-matched wild-types, *, p<0.05, n2:9 animals per h) RPE cells were immunoblotted for beta-tubulin and actin. group. (B) Immunohistochemistry for p62 (green) in retinal 15 (C) Phalloidin staining of the actin cytoskeleton in polarized 1 cryosections from 6-month-old wild-type and ABCA4_ _ primary RPE, control or treated with A2E. (D) Acetylated mice. PR-photoreceptors; RPE-retinal pigment epithe tubulin staining of RPE untreated (control) or treated with lium. Bar: 20 µm. (C) Stills from live imaging of tfLC3 in U18666A (1 µM, 16 h) to induce cholesterol storage (Xu et primary RPE to monitor basal autophagy and autophagic al., 2012). flux after mTOR inhibition by serum starvation or Torin! 20 FIG. 6A-G. Tubulin acetylation modulates autophago- and Torin 2 (both for 2 hours). (D) Quantification ofEGFP/ some trafficking in the RPE. (A) Immunofluorescence mRFP ratios of cells in (C). n=30 cells per condition. *, images of polarized primary RPE, untreated or exposed to significantly greater than corresponding control cells, A2E and stained for acetylated tubulin (green) and ZO-1 p<0.01. (E) Representative immunoblot of LC3B-I, LC3B (red) or tyrosinated tubulin (green) and phalloidin to label II and p62 protein levels in control or A2E-laden primary 25 actin (red). (B) Relative intensities of acetylated and tyro RPE monolayers untreated, or treated with sinated tubulin staining in RPE, Mean±S.E.M., *, p<0.0001 Torins±Bafilomycin Al (Baf, 100 nM for 2 hours). (F) relative to corresponding controls. (C) Immunohistochem Quantification ofLC3B-II immunoblots, n2:9 per condition; istry for acetylated tubulin (red) in retinal cryosections from 1 *, p<0.01; n.s.-not significant. (G) Quantification of p62 6-month-old wild-type and ABCA4_ _ mice. PR-photore- immunoblots, n2:9 per condition; **, p<0.001; n.s.-not 30 ceptors; RPE-retinal pigment epithelium. Bar: 20 µm. (D) significant. Cells stained for acetylated tubulin (green) and ZO-1 (red) FIG. 2. Autophagosome biogenesis in the RPE. Immuno after nocodazole (33 µM) and cold treatment for 30 min. (E) fluorescence staining for WIPI2 (green), ZO-1 (red) and Number of acetylated microtubules longer than 5 µm after DAPI (blue) in control RPE or exposed to A2E as indicated, nocodazole and cold treatment. Mean±S.E.M., *, p<0.0001. all treated with Torin 1 (50 nM) and Torin 2 (1.5 µM) for 2 35 (F) Still from live imaging of EGFP-LC3 trafficking with h. spots and tracks superimposed in primary RPE treated with FIG. 3A-E. Live imaging of autophagosome trafficking in the HDAC6 inhibitor trichostatin A (TSA, 500 nM). (G) the RPE. (A) & (C) Stills from live imaging of EGFP-LC3 Track displacement of EGFP-LC3-labeled autophagosomes with spots and tracks superimposed in serum-starved RPE versus total track length in TSA-treated cells in (F) analyzed (A) untreated (control) or (C) exposed to A2E. Right pan 40 by Boolean gating (See Table 2 for values). els-high-magnification images of areas denoted by white FIG. 7A-D. BMP-mediated activation of ASMase pro boxes in the left panels. (B) & (D) Plots of track displace motes tubulin acetylation. (A) Immunofluorescence images ment of EGFP-LC3-labeled autophagosomes versus total of polarized primary RPE, untreated (Control) or exposed to track length in untreated cells (B) or cells exposed to A2E A2E as indicated, labeled with filipin (blue) and stained for (D). Boolean gating was used to analyze data as explained 45 BMP (red) and ZO-1 (purple). (B) ASMase activity in in the Methods. See Table 2 for Rl, R2 and R3 percentages. primary RPE untreated (control) or treated with A2E and/or (E) Frequency histogram ofEGFP-LC3 labeled autophago the AS Mase inhibitor Desipramine (Des, 10 µM for some velocities (µm/sec). Grey bars-control cells; black 3 hours). *, p<0.005 relative to all other conditions, (one bars-cells treated with A2E. way ANOVA, Dunnett's post-test). (C) Immunofluorescence FIG. 4A-G. Cholesterol removal restores autophagosome 50 staining for ceramide (red) and ZO-1 (white) in untreated biogenesis and autophagic flux in RPE with lipofuscin cells ( control) or cells treated with A2E and/or desipramine. bisretinoids. (A) Biochemical quantification of cholesterol (D) Immunofluorescence staining for acetylated tubulin in polarized primary RPE, untreated (control) or exposed to (green) and ZO-1 (red) in untreated cells (control) or cells A2E as indicated. Mean±S.E.M., *, p<0.005 relative to treated with A2E and/or desipramine. controls. (B) Total RPE cholesterol in eyecups from wild- 55 FIG. SA-E. Desipramine corrects autophagosome traf- 1 type andABCA4_ _ mice. Mean±S.E.M., n2:9 per group.*, ficking defects in the RPE. (A) Stills from live imaging of p<0.005 relative to age-matched wild-types. (C) Represen EGFP-LC3 with spots and tracks superimposed in serum tative immunoblot of LC3B-I, LC3B-II and p62 protein starved RPE treated as indicated. Des, desipramine. (B-E) levels in control or A2E-laden primary RPE monolayers. Analyses of live imaging data depicted as track displace- Cells were untreated or treated with Torins, bafilomycin 60 ment of EGFP-LC3-labeled autophagosomes versus total (Baf, 100 nM for 2 hours) and/or the LXRa agonist track length in control cells and cells treated with A2E TO901317 (1 µM, 20 hours) as indicated. (D) Quantification and/or desipramine. Boolean gating was used to analyze ofLC3B-II immunoblots, n2:9 per condition;*, p<0.05; **, data as explained in the Methods. See Table 3 for Rl, R2 and p<0.01; n.s.-not significant. #, significantly greater than R3 percentages. corresponding condition without TO901317 (in red hatched 65 FIG. 9A-E Inhibition of ASMase activity restores bars), p<0.05. (E) Quantification of p62 immunoblots, n2:9 autophagic flux in RPE with the lipofuscin bisretinoid A2E. per condition; *, p<0.05; n.s.-not significant. #, signifi- (A) Representative immunoblot of LC3B-I, LC3B-II and US 10,124,008 B2 5 6 p62 protein levels in control or A2E-laden primary RPE alone, p<0.001; ASMase inhibitors tested include: astem monolayers. Cells were untreated or treated with Torins, izole (Ast), sertraline (Ser), desipramine (Des), and fex bafilomycin (Baf, 100 nM for 2 hours) and desipramine ofenadine (Fex). (Des, 10 µM, 3 h) as indicated. (B) Quantification of LC3B-II immunoblots, n2:9 per condition; **, p<0.01; n.s.- 5 DETAILED DESCRIPTION OF THE not significant. #, significantly greater than corresponding INVENTION condition without desipramine (in red hatched bars), p<0.01. (C) Quantification of p62 immunoblots, n2:9 per condition; We have identified a stepwise mechanism by which **, p<0.01; n.s.-not significant.#, significantly lesser than excess lysosomal cholesterol traps a lipid called bis(monoa corresponding condition without desipramine (in red 10 cyl-glycero )phosphate (BMP) within lysosomes. BMP is an hatched bars), p<0.001. (D) Stills from live imaging of activator of ASMase activity and the resulting increase in mRFP-GFP-LC3 and quantification of EGFP/mRFP ratios ceramide levels interfere with autophagosome biogenesis, in serum-starved RPE treated as indicated. **, significantly autophagosome traffic and fusion with lysosomes. Declining greater than all other treatments, p<0.0001. (E) Stills from autophagy with age has emerged as a key pathogenic factor live imaging of tf-LC3 and quantification of EGFP/mRFP 15 in Alzheimer's and Parkinson's diseases, and could also ratios in torin-treated RPE treated as indicated. *, signifi cantly greater than all other treatments, p<0.01. contribute to RPE dysfunction as a precursor to permanent FIG. lOA-C. Model for impaired autophagy in RPE with vision loss. bisretinoids. (A) Cone-shaped bisretinoids like A2E (#1) This finding suggests to us that inhibition of ASMase is a sequester cholesterol (#2) in RPE late endosomes and lyso- 20 therapeutic target in retinal dystrophies associated with sames, as we have reported previously (Lakkaraju et al., lipofuscin accumulation. Thus, tricyclic antidepressants and 2007). Data presented in this study show that excess cho other ASMase inhibitors constitute a new line of therapy for lesterol in turn traps BMP (#3), which activates ASMase AMD and for inherited retinal diseases characterized by (#4), leading to increased production of ceramide from abnormal accumulation of lipofuscin. sphingomyelin (#5). Ceramide promotes tubulin acetylation 25 After we made the connection between ASMase inhibi on stable microtubules (#6) possibly by inhibiting HDAC6. tion and retinal disease treatment, we examined two epide Acetylated microtubules in the RPE interfere with miological studies, the Beaver Dam Eye Study based at UW autophagosome motility (#7). (B) In normal RPE, (Klein et al., 2001) and a pooled study from three continents autophagosome biogenesis and transport are essential for (van Leeuwen et al., 2004), which indicate that use of efficient autophagic flux. (C) Our data show that in RPE with 30 tricyclic antidepressants (TCA) is associated with a bisretinoids aberrant activation of ASMase interferes with decreased risk of developing early AMD. While these initial multiple steps of autophagy by increasing ceramide-induced studies on TCA use in AMD were published over a decade tubulin acetylation. Whether acetylated microtubules also ago, they did not address the mechanism underlying the interfere with the trafficking of other organelles like phago protective effect (Klein et al., 2001; van Leeuwen et al., somes, endosomes and lysosomes is currently under inves- 35 2004). These studies had no baseline data on dosage and tigation. Drugs that inhibit ASMase decrease tubulin acety duration of medication use, and both one-time users and lation and increase autophagy in cells with lipofuscin chronic users of TCAs were grouped together. bisretinoids (#8). In summary, ASMase inhibitors are effective for both FIG. llA-E. (A) Expression of GPI-anchored comple preventing and treating macular dystrophies because (i) our ment regulatory proteins CD55 and CD59 (both green) in 40 data show that a short 3 hour incubation with desipramine, 1 retinal cryosections from wild-type or Abca4_ _ mice. (B) sertraline or astemizole completely reversed the autophagic Surface CD59 staining in polarized primary RPE untreated block induced by A2E; (ii) other ASMase inhibitors ami (control), or exposed to A2E or A2E and desipramine (des). triptyline, fluoxetine and zoledronic acid decreased acety (C) Surface LAMP2 (green) in primary RPE monolayers lated tubulin levels in RPE withA2E back to control levels; exposed to 10% normal human serum (NHS) as a source of 45 (iii) Desipramine maintained RPE integrity and prevented complement. The tight junction protein ZO-1 is in red. (D) mitochondrial oxidative damage after complement attack; Quantification of reactive oxygen species in primary RPE (iv) ASMase inhibitors are low molecular weight, lipophilic monolayers exposed to 10% NHS. Beige bars----control drugs that are well-absorbed after oral administration and RPE; red bars-cells loaded with A2E prior to complement are blood-brain barrier permeable, which makes it likely that attack. E-Quantification of MHC-II staining. *, p<0.001. 50 they will be permeable to the RPE outer blood-retinal barrier FIG. 12A-C. (A) shRNA-mediated knockdown of and the corneal barrier after topical administration; and (v) AS Mase (SMPD 1 gene) in primary RPE. Construct D is the these drugs are all FDA approved and have well-established most effective. (B) Immunostaining for ASMase (green) and safety and efficacy profiles. acetylated tubulin (white) in cells expressing shRNA to ASMase (red). Yellow asterisk denotes transfected cells in 55 Diseases Treated by the Present Invention which ASMase has been knocked down. White # indicate untransfected cells that served as internal controls for com In one embodiment, the present invention is a method of parison. (C) LC3-labeled autophagosomes per cell after treating retinal diseases comprising the steps of administer scrambled or SMPDl shRNA transfection. *, significantly ing an effective amount of a composition comprising an lower than all other conditions, p<0.05. 60 ASMase inhibitor to a retinal disease patient, wherein dis FIG. 13A-C. (A) Acetylated tubulin (green) staining in ease symptoms are lessened. cells treated with A2E and ASMase inhibitors desipramine Preferred diseases for treatment by the present invention (des), amitriptyline (ami), fluoxetine (fluo) or zoledronic include age-related macular degeneration (AMD), Stargardt acid (ZA). All 10 µM for 3 h, 48 h after A2E treatment. (B) macular dystrophies (autosomal dominant and autosomal Quantification of acetylated tubulin fluorescence in A. (C) 65 recessive forms), Best vitelliform macular dystrophy and EGFP:mRFP ratios as a measure of tfLC3 flux in primary neuronal ceroid lipofuscinoses (NCLs), such as Batten's RPE. *, significantly lower than cells treated with A2E Disease. These diseases are all associated with an abnormal US 10,124,008 B2 7 8 accumulation of lipofuscin, a complex mixture of visual Nature 2012), which are the five physicochemical features cycle retinoids, in the retinal pigment epithelium (RPE) that predict oral bioavailability of the drug such as pKa, mo! (Travis et al., 2007). wt, lipophilicity, etc. One would preferably administer these The method is suitable for emerging disease states. If a drugs using eye drops. For example, drugs may be dissolved patient is deemed at risk of developing early AMD by their 5 in sterile-filtered aqueous-based solutions containing 1% ophthalmologist after a fundus exam, the patient could likely hydroxypropyl methylcellulose, 0.2% tyloxapol, 3.4% dex be a candidate for treatment. One would most likely start treatment when the exam reveals presence of soft, large trose, 0.006% benzalkonium chloride, and 0.025% ethyl drusen, which are confirmed risk factors for AMD. Treat enediaminetetraacetic acid (280 mOsm, pH 5.4), which has ment is suitable for any age/gender. been shown to deliver small molecular weight compounds to 10 the retina, RPE and choroid (Doukas et al., J Cell Physiol, ASMase Inhibitors of the Present Invention 2008; Ganrmons et al., IOVS 2013) The present invention includes the use of an acid sphin 2. Liposomes gomyelinase (AS Mase) inhibitor for the treatment of acquired and inherited retinal diseases. In our studies 15 Drugs may be encapsulated in liposomes and subjected to described below, we tested inhibitors of ASMase (Kaizer et microfluidization, which decreases Ii po some size to <60 nm. al., 2004), and desipramine (a functional inhibitor) is a preferred ASMase of the present invention. Desipramine is Liposomes of this size administered topically ( directly to the also known as desmethylimipramine. Brand names are eye) can reach the RPE (Lajunen et al., 2014). Liposomes NORPRAMIN and PERTOFRANE. Desipramine is the 20 may be composed of phospholipids such as phosphatidyl active metabolite of imipramine (also known as melipramine choline, phosphatidylserine phosphatidylethanolamine and and sold under the brand name TOFRANIL). coated with polyethylene glycol to increase tissue stability. We envision that other ASMase inhibitors would be Preferably, liposomes will be actively targeted to the RPE suitable. A number of low molecular weight cationic lipo using either the transferrin receptor (Lajunen et al., 2014) or philic drugs that are currently approved by the FDA and are 25 the low-density lipoprotein-related receptor protein (LRP) on the market are known to inhibit ASMase function (see (Lakkaraju et al., 2002). Table 1 of Kornhuber et al., 2010, incorporated by refer ence). These drugs are all orally bioavailable and have 3. Gels and Ointments established safety and efficacy profiles. We envision that these drugs will be suitable topically as well as orally. The best orally bioavailable drugs are small 30 In one embodiment of the invention, the inhibitor is molecular weight, lipophilic compounds which are also delivered in a gel or ointment delivered directly to the eye. ideal characteristics for topical eye formulations. For example, drugs may be loaded in bases such as CAR Additionally, we envision that drugs that are known to BOPOL 934 or liquid paraffin listed in the US Pharmaco structurally inhibit ASMase (e.g., bisphosphonates like zole poeia. dronic acid (Roth et al., 2009)) would be suitable. 35 We further envision that additional dibenzazepine deriva Representative Dosages and Dosage Schedules tives aside from desipramine may be suitable. If one were to apply the medication as a topical applica Preferred Methods of Delivery of Acid tion, we envision that the dose would be applied daily or Sphingomyelinase Inhibitors to the Retinal Pigment 40 every other day until efficacy is established. Currently, we Epithelium use 2.6 µg/ml in vitro. Assuming 95% of topically applied dose doesn't reach target (usual case with eye formulations), We envision that acid sphingomyelinase inhibitors will be we envision starting at 0.1 mg and going up to 1 mg per delivered orally (either single drugs or in combination), dose. topically or via sustained release scleral implants. 45 If one were to use an oral administration, doses will be Applicants envision that the treatment method of the present invention may be of different modalities. For based on published data for the inhibitor (Gulbins et al., example, one may wish to deliver the ASMase inhibitor Nature Medicine, 2013 and Petersen et al., Cancer Cell, orally, intravenously, intravitreal injections, topical, trans 2013). Generic oral dosages used in patients currently: mucosal (nasal, buccal/sublingual, vaginal, ocular and rec 10-25 mg/day norpramin (desipramine); 50-100 mg/day tal) and inhalation routes. Note that topical delivery directly 50 tofranil (imipramine ). We envision that a preferred oral dose to the eye may be via an ocular route, which can be range will be between 10-100 mg/day. transmucosal because the drug is thought to go through the mucous membrane of the conjunctiva. We also envision that Evaluating Effective Drug Delivery the drug may be delivered through corneal drug delivery, which is not transmucosal. 55 After the treatment of the present invention, one will wish In one version of the invention, the drug is delivered to evaluate the efficacy of the treatment. In one embodiment, orally. Generic oral dosages used in patients currently are one may evaluate the progression of the disease. In one 10-25 mg/day NORPRAMIN (desipramine) and 50-100 embodiment, a successful treatment would result in lack or mg/day tofranil (imipramine). slowing of disease progression. For example, for an AMD In preferred embodiments of the present invention, we 60 patient, one would wish to evaluate the progression of the envision to following preferred formulations for topical drug following disease symptoms: accumulation of soft, large delivery: drusen; increased fundus autofluorescence; delayed dark adaptation. 1. Eye Drop Solutions If one were evaluating the treatment from a molecular 65 level (for example, with a disease model), one may wish to Many of the acid sphingomyelinase inhibitors we have examine: 1) Quantification of ceramide levels and acid disclosed conform to the Lipinski rule of five (Leeson, sphingomyelinase activities in the RPE after topical and oral US 10,124,008 B2 9 10 administration in mice using mass spectrometry; and 2) the RPE for life because their unique structures render them Measurement of autophagic flux in the RPE in wild-type and resistant to lysosomal degradation. Stargardt disease mice. Chronic accumulation of these bisretinoids has been implicated in the pathology of numerous blinding retinal EXAMPLES 5 diseases including Stargardt disease, Best disease and age related macular degeneration (AMD) (Ambati and Fowler, Example 1. Cholesterol-Mediated Activation of 2012; Sparrow et al., 2012). We previously demonstrated Acid Sphingomyelinase Disrupts Autophagy in the that A2E, a cone-shaped lipid, displaces cholesterol from Retinal Pigment Epithelium lipid bilayers and sequesters cholesterol in RPE late endo- 10 sames and lysosomes (Lakkaraju et al., 2007). Since mem Macroautophagy (hereafter referred to as autophagy) is a brane cholesterol levels modulate autophagosome-lysosome bulk degradative pathway where double-membraned struc interactions (Fraldi et al., 2010; Koga et al., 2010; Sarkar et tures called autophagosomes enclose damaged proteins and al., 2013), we hypothesized that cholesterol storage induced organelles. Fusion of autophagosomes with the endo-lyso by lipofuscin bisretinoids would inhibit autophagic clear- 15 ance in the RPE. OS phagocytosis has been shown to recruit somal system delivers hydrolytic enzymes required to autophagic machinery in the RPE (Kim et al., 2013; Frost et degrade the sequestered cytosolic components (Rubinsztein al., 2014; Yao et al., 2014) and inefficient autophagy is et al., 2007; Choi et al., 2013). Formation of the autophago thought to play a part in the pathogenesis of retinal diseases some is initiated by the activation of autophagy-related such as AMD (Bowes Rickman et al., 2013; Frost et al., (Atg) proteins in a hierarchical manner. The molecular 20 2014). However, how innate processes such as progressive machinery of autophagy is highly conserved and primarily accumulation of lipofuscin bisretinoids impact autophagy in regulated by the mammalian target of rapamycin (mTOR), the RPE is not well understood. in response to the nutrient and metabolic status of the cell. Here, we report decreased autophagosome biogenesis and 1 Autophagy occurs at a basal level in most cells and is autophagic flux in the RPE of ABCA4_ _ disease mice, increased under conditions of stress, when it promotes 25 which have high levels of A2E and other bisretinoids (Radu survival by repurposing degraded material to support et al., 2011 ). High-speed live imaging of primary RPE by metabolism within the cell (Codogno et al., 2012). spinning disk confocal microscopy (Toops et al., 2014) Inefficient autophagy has been implicated in the patho showed that A2E interfered with autophagosome biogenesis, genesis of neurodegenerative diseases because post-mitotic constrained autophagosome traffic and decreased autophagic neurons are especially susceptible to the accumulation of 30 flux. Our data unveil a step-wise molecular mechanism by defective organelles and protein aggregates (Nixon, 2013). which lipofuscin- and A2E-induced lysosomal cholesterol Autophagy is also critical for maintaining the health of the storage (Lakkaraju et al., 2007) activates acid sphingomy neural retina: in aged mice or mice with retina-specific elinase (AS Mase) by sequestering the anionic lipid bis deletions of Atg5, decreased autophagic flux precedes pho (monoacyl-glycero )phosphate (BMP), an ASMase co-factor toreceptor degeneration (Rodriguez-Muela et al., 2013) and 35 (Kirkegaard et al., 2010). The resulting increase in ceramide autophagy induction preserves differentiation of the retinal levels lead to increased tubulin acetylation (He et al., 2012; pigment epithelium (RPE) and prevents photoreceptor death He et al., 2014). Our data show that bidirectional motility of after oxidative stress (Zhao et al., 2011) or exposure to Fas autophagosomes and autophagosome-lysosome fusion are ligand (Besirli et al., 2011 ). In the post-mitotic RPE, which impaired in cells with acetylated microtubules. In support of nourishes and supports the overlying photoreceptors, 40 a central role for cholesterol-mediatedASMase activation in autophagy is increased in response to diverse stressors regulating autophagy, we demonstrate that a drug that pro- including exposure to intense light, oxidative stress, mito motes cholesterol efflux (Lakkaraju et al., 2007) and an US chondrial poisons, cigarette smoke and cell swelling (Reme Food and Drug Administration (FDA)-approved ASMase et al., 1999; Kunchithapautham and Rohrer, 2007a, b; Chen inhibitor (Kornhuber et al., 2010) restore efficient et al., 2013; Doyle et al., 2014; Wang et al., 2014). 45 autophagosome transport and autophagic flux in the RPE. A key function performed by the RPE critical for photo There are two significant implications of our study: first, receptor health is the circadian phagocytosis and lysosomal our data show that autophagy in the RPE is regulated in degradation of shed photoreceptor outer segment (OS) tips response to the immense metabolic demands placed on the (Bok, 1993). Each RPE cell contacts 30-50 photoreceptors, cell, adding to a growing body of evidence for specialized which shed about 10% of their OS length daily. Over a 50 regulation of autophagy based on tissue, function and con lifetime, this immense metabolic activity results in the text (Grumati et al., 2010; Jimenez-Sanchez et al., 2012; Le progressive accumulation of undigested OS components Guezennec et al., 2012; Pampliega et al., 2013). Second, our called lipofuscin in RPE lysosomes (Sparrow et al., 2012). studies suggest that ASMase inhibition could be a potential RPE lipofuscin differs from that in other post-mitotic tissues novel therapeutic strategy not only in macular degenerations in that it is primarily composed ofbisretinoid metabolites of 55 associated with excess lipofuscin accumulation but also in vitamin A, generated as by-products of the visual cycle diseases characterized by abnormal cholesterol homeostasis (Eldred and Lasky, 1993). Light induces isomerization of the and impaired autophagy (Le Guezennec et al., 2012; Nixon, visual chromophore 11-cis-retinal (11 CR) to all-trans-retinal 2013; Barmada et al., 2014; Lee et al., 2014). (ATR), which is flipped by the ATP binding cassette trans porter A4 (ABCA4) from the lumen to the cytosolic side of 60 Materials and Methods the disc membrane and reduced to non-toxic all-trans-retinal by retinal dehydrogenase 8 (Weng et al., 1999). Delayed Cells removal of 11 CR and ATR from disc membranes makes them susceptible to condensation reactions that ultimately Primary RPE were harvested from freshly enucleated result in the formation of vitamin A derivatives such as the 65 porcine eyes (Hart and Vold, Baraboo, Wis.) as described lipofuscin bisretinoid A2E in RPE lysosomes (Sparrow et (Toops et al., 2014): briefly, the anterior segment was al., 2012). Once formed, lipofuscin bisretinoids remain in removed at the ora serrata and the retina was gently detached US 10,124,008 B2 11 12 by clipping at the optic nerve head. RPE cells were isolated followed by cold-treatment ( 4 ° C.) for 30 minutes (Kreitzer from eyecups upon incubation with 0.5% trypsin with 5.3 et al., 2003). At the concentrations and exposure times used, mM EDTA in HBSS and plated onto T25 flasks in DMEM none of these drugs caused alterations in RPE cell morphol with 1% heat-inactivated fetal bovine serum (FBS, ATCC). ogy or physiology (monitored by TER measurements, ZO-1 To generate polarized cultures, cells were plated at conflu- 5 and organelle marker staining). 2 ence (-300,000 cells/cm ) onto collagen-coated TRAN SWELL (Corning) semi-permeable membrane filters. After Immunofluorescence Staining and Quantification two weeks, monolayers had trans-epithelial electrical resis 2 tances of >300 ohm·cm , localized Na+, K+-ATPase api Filter-grown cells were fixed in 2% paraformaldehyde for cally, expressed tight junction proteins ( e.g., ZO-1) and RPE 10 10 min, blocked in 1% BSA in PBS and incubated with differentiation markers (e.g., RPE65) (Toops et al., 2014). specific primary antibodies for 1 hour: mouse monoclonal anti-acetylated tubulin clone 6-118-1 (1:1000, Sigma), rat Animals monoclonal anti-tyrosinated a-tubulin (1:200, Santa Cruz), mouse anti-LBPA (1:500, Echelon Z-LBPA), mouse anti- 1 Wild-type and ABCA4_ _ mice (both 129/Sv strain on 15 ceramide (1:10, Enzo) and rat anti-Z0-1 1:3000 (Xu et al., Rpe65 Leu450 background) were raised under a 12-h cyclic 2012). ALEXAFLUOR secondary antibodies were used at light and fed a standard rodent diet (NIH-31, 7013 Harlan 1:500 and rhodamine-phalloidin (Cytoskeleton, PHDRl) at Teklad, Madison, Wis.). Mouse studies were done in adher 1 :200. Filters were mounted under coverslips on glass slides ence to guidelines established by the UCLA Animal under VECTASHIELD (Vector labs), sealed and visualized Research Committee and The Association for Research in 20 with Andor Revolution XD spinning disk confocal micro Vision and Ophthalmology statement for the Use ofAnimals scope using a 60xl.4 NA oil objective with identical expo in Ophthalmic and Vision Research. Animals were eutha sures and gains for each antibody. Acetylated tubulin-la nized -4-6 h after light onset, eyes were removed and beled stable microtubules (;;,;5 µmin length) after nocodazole hemisected. The anterior portion containing the cornea, lens, treatment were analyzed manually in the Surpass mode of and vitreous was discarded. Eyecups containing retina, RPE, 25 Imaris (Bitplane). choroid, and sclera were frozen in liquid N2 and stored at -80° C. for further processing (Radu et al., 2011). Immunohistochemistry 1 Immunoblotting Cryosections of wild-type and ABCA4_ _ mice retinas 30 were blocked in PBS with 4% BSA and incubated with RPE harvested from mouse eyecups were sonicated in primary antibodies (diluted 1:100 in PBS with 4% BSA) for lysis buffer with protease inhibitors for 10 min. Primary RPE 48 hat 4° C. in a humidified chamber. Slides were rinsed to on TRANSWELL filters were harvested and lysed with remove unbound antibodies and incubated with Alexa-con- NE-PER nuclear and cytoplasmic extraction reagents jugated secondary antibodies (1 :500 in PBS with 4% BSA) (Thermo Scientific #78833) according to the manufacturer's 35 for 18 hat 4° C. in a humidified chamber protected from recommendation. Protein concentrations were measured light. Sections were rinsed, stained with DAPI for 5 min, with DC assay (BioRad). Samples (20 µg/lane) were rinsed and sealed under coverslips using VECTASHIELD as resolved in 4-12% NUPAGE Bis-Tris Precast Gels (INVIT a mounting medium. Slides were imaged with the Andor ROGEN) at 130V. Proteins were then transferred onto Revolution XD spinning disk confocal microscope using a nitrocellulose membrane using IBLOT dry transfer system 40 40xl .4 NA oil objective with identical exposures and gains (INVITROGEN), blocked in 5% milk in TBS-T for 1 hour for each antibody. before incubating in primary antibody overnight at 4 ° C. Membranes were probed with antibodies to LC3B (1 :3000, Transfections Novus NB600-1384 for pig and 1:500, Sigma L7543 for mouse), p62/SQSTM1 (1:1000, ARP03-GP62-C) and actin 45 RPE cells were transfected with EGFP-LC3 or tandem (1:5000, Santa Cruz) followed by horseradish peroxidase fluorescent mRFP-GFP-LC3 (Addgene) using the AMAXA conjugated secondary antibodies. Immunoblots were visu NUCLEOFECTOR II (Lonza). Approximately 1.5 million alized by ECL substrate (Thermo Scientific) and quantified cells and 5 µg of plasmid DNA were used for each trans using Image Studio (LI-COR). fection. Cells were plated either on serum-coated glass- 50 bottom dishes (Mattek) or TRANSWELL filters at conflu- Pharmacological Treatments ence. The lipofuscin bisretinoidA2E was synthesized according Spinning Disk Microscopy to published protocols and purified by HPLC (>97%, ESI MS) (Lakkaraju et al., 2007). RPE were exposed to either a 55 Live imaging of autophagosome traffic and autophagic chronic low-dose of A2E (50 nM for 3 weeks) or an acute flux were performed on the Revolution XD spinning disk high-dose of A2E (10 µM for 6 h, followed by a 48 h chase). microscopy system (Andor) equipped with the Yokogawa Quantification of A2E levels in cells was performed by CSU-XI confocal spinning disk head, Nikon Eclipse Ti HPLC as previously reported (Radu et al., 2011 ). Other inverted microscope surrounded by an Okolab cage incuba drugs used were the mTOR inhibitors Torin 1 and Torin 2 60 tor, iXon x3 897 EM-CCD camera, Andor laser combiner (50 nM and 1.5 µM, respectively, for 2 h, TOCRIS), the with four solid state lasers at 405, 488, 561, and 640 nm and vacuolar ATPase inhibitor bafilomycin Al (100 nM for 2 h, corresponding band-pass filter sets (Sutter), and ASI motor EMD Millipore), the LXRa agonist TO901317 (1 µM for 20 ized stage with piezo-Z for rapid Z-stack acquisition. Andor h, Cayman Chemicals), the HDAC6 inhibitor trichostatin A IQ2 software was used for image acquisition and Imaris x64 (TSA, 500 nM, 16 h, Sigma) and the ASMase inhibitor 65 (Bitplane) for image analysis. For live imaging, cells were desipramine (10 µM for 3 h, Sigma). To depolymerize MTs, serum starved for 2 h to induce autophagy and rapid z-stacks cells were treated with 33 µM nocodazole for 30 minutes, were acquired using the l00xl.49 NA Apo TIRF objective US 10,124,008 B2 13 14 (Nikon) for -50 frames at 37° C. Trafficking data was Regions were defined by: minimum or maximum track collected from three separate transfections for a total of at length, minimum or maximum displacement and a slope of least 12-30 movies captured for treatment. During image 0.5 (displacement divided by length). Tracks with slope <0.5 acquisition care was taken to maintain the same laser power, were taken as less straight (region 2) than those with slope exposure and electron-multiplying gain settings. Trafficking 5 of>0.5 (region 3). Tracks within each region are represented analysis was carried out using the Spots and Tracks modules as a percent of total number of tracks in Tables 1 and 2. of the Imaris software (Liu et al., 2010). After background One-way ANOVA or t-tests were used to compare regions subtraction (using the background subtraction algorithm and between treatment groups. identical automatic threshold for all images) and smoothing (Gaussian algorithm with identical threshold settings for all 10 images), Spots and Tracks algorithms were used to identify Results vesicles and follow them through time and cell-space to obtain total track length and track displacement. Statistical analysis of these data sets was performed using Excel Lipofuscin Bisretinoids Interfere with Canonical Autophagy in the RPE (Microsoft) and PRISM (GraphPad). 15 Biochemical Assays To investigate whether lipofuscin bisretinoids impact autophagy in vivo, we measured microtubule-associated Cells were harvested by trypsinization and cell pellets light chain 3B-II (LC3B-II) and p62/SQSTM1 levels in RPE 1 were washed with PBS to remove residual medium. Cells 20 of ABCA4_ _ mice, which have high levels of lipofuscin were lysed in HNTG lysis buffer (50 mM Hepes, pH 7.4, 150 bisretinoids such as A2E (Radu et al., 2011). Conversion of mM NaCl, 10% glycerol, 1.5 mM MgC1 , 1% triton X-100) 2 LC3B-I to its lipidated form (LC3B-II) is an indicator of supplemented with protease inhibitors. Total protein was autophagosome biogenesis and p62 levels are a measure of measured using the DC protein assay kit (BioRad). Cellular 1 autophagic flux (Klionsky et al., 2012). ABCA4_ _ RPE had cholesterol was quantified using the Amp lex Red cholesterol 25 assay kit (INVITROGEN) as detailed previously (Lakkaraju significantly less LC3B-II (FIG. lA) and more p62 (FIG. 1 et al., 2007). Single eyecups from wild-type and ABCA4_ _ lB) compared to RPE from age-matched wild-type mice, mice were homogenized in 50 µl of lysis buffer and pro supporting the hypothesis that accumulation of lipofuscin cessed as above. For measuring acid sphingomyelinase bisretinoids is associated with defective autophagy in vivo. activity, cells were lysed in acidic pH and assays were 30 To determine how lipofuscin bisretinoids inhibit autophagy performed using the sphingomyelinase fluorimetric assay kit in the RPE, we established an in vitro model by exposing from Cayman Chemicals according to the manufacturer's polarized porcine primary RPE monolayers (Toops et al., protocol. 2014) to the bisretinoid A2E, chronically ( 50 nM over three weeks) or acutely (10 µM for 6 h), either of which result in Statistical Analysis 35 intracellular A2E levels comparable to those seen in the RPE 1 Data were analyzed using either a two-tailed t-test or of ABCA4_ _ mice and in human Stargardt disease patients one-way ANOVA followed by Bonferroni or Dunnett's (Table 1). TABLE 1 A2E levels in polarized primary pig RPE, mouse models and humans pmoles A2E per 100,000 ng A2E per RPE 100,000 RPE Model Details (mean± SD) (mean± SD) Notes Primary pig RPE No A2E control Not detectable Toops et al. 15 nM, 3 weeks 4.92 ± 1.44 2.91 ± 0.85 50 nM, 3 weeks 14.13 ± 2.71 8.37 ± 1.61 10 µMA2E, 6 h 22.32 ± 0.23 13.21 ± 0.14 Wild-type mice 3 months old 1.58 ± 0.34 0.93 ± 0.20 (Radu et al., 2011) 6 months old 3.32 ± 0.01 1.96 ± 0.006 12 months old 9.30 ± 1.80 5.51 ± 1.06 Abca4_1_ mice 3 months old 8.37 ± 0.54 4.96 ± 0.32 (Radu et al., 2011) 6 months old 23.32 ± 2.47 13.81 ± 1.46 12 months old 28.30 ± 5.80 16.75 ± 3.45 Normal human Seven normal 5.30 ± 2.98* 3.14 ± 1.76 * per 0.25 cm2 RPE subjects (Mata et al., 2000) Stargardt's STGDl #1 33* 19.54 * per 0.25 cm2 human RPE STGDl #2 61 * 36.11 (Mata et al., 2000) Aging human Ten subjects, 145.27 ± 54.17 86 ± 32.1 ( Sparrow et al., RPE (normal) 58-79 years 1999) post-tests (GraphPad PRISM). Unless otherwise stated, data In polarized primary RPE cells, autophagy was upregu are presented as Mean±S.E.M. of 2:3 independent experi lated after mTOR inhibition, either by nutrient deprivation ments, with 3 to 4 replicates per condition per experiment. or treatment with the selective mTOR inhibitors Torin 1 and To analyze EGFP-LC3 trafficking data, Boolean gating was 65 Torin 2. We used tandem fluorescent mRFP-GFP-LC3 used to segment three regions (denoted by Rl, R2 and R3 on (tfLC3) to monitor autophagy in primary RPE cells in real the track displacement versus total track length graphs). time. Upon fusion of autophagosomes with lysosomes, US 10,124,008 B2 15 16 EGFP fluorescence is quenched in the acidic lysosomal pH TABLE 2 and only the pH-insensitive mRFP signal is visible; there fore, ratios of EGFP to mRFP are a measure of autophagic Ouantitation of EGFP-LC3 trafficking data (from FIGS. 3 and 6) flux (Klionsky et al., 2012). TfLC3 imaging showed signifi Region 2 (R2) Region 3 (R3) cantly more EGFP puncta in cells treated with A2E com- 5 Region 1 (Rl) (L >5 µrn, (L >5 µrn, pared to control cells, both at the basal level and after mTOR (D <2 µrn, L <5 µrn) Slope <0.5 µrn) Slope >0.5 µrn) inhibition either by nutrient deprivation or Torin treatment Control(%) 87.57 6.56 5.87 (FIGS. lC, lD). A2E (%)* 93.51 5.37 1.12 TSA (%)* 88.01 10.35 1.64 Immunoblotting of polarized primary RPE monolayers 10 after mTOR inhibition showed significantly lower LC3B-II *significantly different from control cells, p < 0.0001, one-way ANOVA. levels in cells treated with A2E compared to control RPE. The vacuolar ATPase inhibitor Bafilomycin Al, which pre Excess Cholesterol Mediates Autophagic Defects in vents lysosomal degradation of LC3B, increased LC3B-II RPE with Bisretinoid after mTOR inhibition in control cells but not cells exposed 15 to A2E (FIGS. lE, lF). To confirm that A2E decreased A critical determinant of organelle motility and fusion is autophagosome biogenesis, we immunostained primary membrane cholesterol (Lebrand et al., 2002; Fraldi et al., RPE monolayers for WD repeat domain, phosphoinositide 2010). We showed previously thatA2E, a cone-shaped lipid, competes with cholesterol (another cone-shaped lipid) for interacting 2 (WIPI2)-positive preautophagosomal struc 20 space under the phospholipid umbrella to minimize unfa tures (Polson et al., 2010). There were fewer WIPI2-labeled vorable interactions with the aqueous phase. Displacement nascent autophagosomes at the basal state and after mTOR of cholesterol from the lipid bilayer traps cholesterol within inhibition in cells withA2E, compared to control cells (FIG. RPE late endosomes and lysosomes (Lakkaraju et al., 2007). 2). Immunoblotting also showed that there was significantly 25 A2E and other lipofuscin bisretinoids increased total cell more p62 in cells with A2E after Torin treatment (FIGS. lE, cholesterol measured biochemically in primary RPE after lG), indicating a block in autophagic flux. Taken together, chronic or acute exposure (FIG. 4A) and in the RPE of 3- 1 these data suggest that lipofuscin bisretinoids interfere with and 6-month old ABCA4_ _ mice (FIG. 4B). We then asked 1 canonical autophagy in vivo in the ABCA4- - mice and in whether A2E-induced cholesterol accumulation was respon- primary RPE cells in culture. 30 sible for autophagic defects. To test this, we treated cells with the liver X receptor alpha (LXRa) agonist TO901317, which transcriptionally activates ABCAl and ABCG 1 cho Autophagosome Trafficking is Disrupted in RPE lesterol transporters, to clear excess cholesterol in cells with with the Bisretinoid A2E A2E (Lakkaraju et al., 2007) (FIG. SA). Immunoblotting 35 and quantification of LC3B-II and p62 protein levels after Since tfLC3 imaging and p62 immunoblotting data mTOR inhibition showed that TO901317 increased autophagosome biogenesis (FIGS. 4C, 4D) and autophagic showed a block in autophagosome-lysosome fusion and flux in A2E-laden cells (FIGS. 4C, 4E). We used live decreased autophagic flux in RPE with bisretinoids, we imaging of tfLC3 to follow autophagosome-lysosome asked whether A2E interfered with the trafficking of 40 fusion: TO901317 restored autophagic flux in cells with A2E autophagosomes. We performed live imaging of EGFP and decreased EGFP fluorescence back to control levels LC3-labeled autophagosomes in untreated or A2E-laden (FIGS. 4F, 4G). Collectively, these data confirm that excess primary RPE cells using high-speed spinning disk confocal cholesterol induces defects in autophagosome trafficking and autophagosome-lysosome fusion in cells with lipofuscin microscopy. After serum starvation to induce autophagy, 45 bisretinoids. RPE with A2E had fewer EGFP-LC3 motile tracks com- pared to control cells (FIGS. 3A, 3C). We used 4D image Tubulin Acetylation on Stable Microtubules Impairs analysis (Imaris, Bitplane) to quantify changes in Autophagosome Trafficking autophagosome trajectories induced by A2E. To examine the efficiency of autophagosome transport, Spots and Tracks 50 Intracellular trafficking is coordinated by the actin and microtubule cytoskeletons and associated motor proteins algorithms in Imaris were used to calculate how far each (Rodriguez-Boulan et al., 2005). Organelle-specific recruit autophagosome traveled (track displacement length) as a ment of microtubule motors is accomplished in part by function of the movement required to travel that distance post-translational modifications of a-tubulin such as acety- (total track length) (Liu et al., 2010). Analysis of track 55 lation and detyrosination, which preferentially occur on displacement data by Boolean gating showed a significant stable microtubules and cause cell type-specific alterations decrease in the population of autophagosomes with long of organelle motility (Joseph et al., 2008; Perdiz et al., range, directed movements (large displacement with long 2011 ). To examine how bisretinoid-induced cholesterol track lengths) in cells withA2E (FIGS. 3B, 3D and Table 2). accumulation interferes with autophagy in the RPE, we Compared to control cells, RPE with A2E had fewer motile 60 asked whether altered microtubule stability and/or post translational tubulin modifications could explain the con autophagosomes (aggregate number of motile tracks was strained trafficking of autophagosomes. Immunostaining 17,010 in control cells and 5,914 in cells with A2E), which showed that acetylated tubulin, which is mainly found in moved with significantly lower velocities (FIG. 3E). How primary cilia of control RPE, increased dramatically in cells could A2E, which is present in RPE late endosomes and 65 with A2E, with a corresponding decrease in tyrosinated lysosomes (Lakkaraju et al., 2007), interfere with tubulin (FIGS. 6A, 6B). Neither total a-tubulin expression autophagosome trafficking? (FIG. SB) nor the organization of the actin cytoskeleton US 10,124,008 B2 17 18 (FIG. SC) were altered in these cells. RPE from 6-month-old desipramine corrected defects in autophagosome-lysosome 1 ABCA4- - mice had more acetylated tubulin compared to fusion after serum starvation (FIG. 9D) and Torin treatment age-matched wild-types (FIG. 6C), confirming that bis (FIG. 9E) in RPE cells with A2E. Thus, ASMase inhibition retinoids increase tubulin acetylation in vivo. Under condi could be an effective therapeutic target to increase cellular tions that depolymerize microtubules in polarized epithelia 5 clearance in RPE with lipofuscin bisretinoids. (nocodazole and cold treatment) (Kreitzer et al., 2003), there were significantly more acetylated microtubules in A2E- TABLE 3 laden cells compared to controls (FIGS. 6D, 6E), indicative of increased microtubule stability. Acetylated tubulin also Quantitation of EGFP-LC3 trafficking data (from FIG. 8B-E) increased in cells treated with Ul 8666A, a drug that induces 10 Region 2 (R2) Region 3 (R3) lysosomal cholesterol storage (Ko et al., 2001) (FIG. SD), Region 1 (Rl) (L >10 µrn, (L >10 µrn, suggesting that cholesterol mediates the effects of A2E on (D <2.5 µrn, L <10 µrn) Slope <0.5 µrn) Slope >0.5 µrn) microtubule stability and tubulin acetylation. To establish Control 85.73 7.57 6.7 that hyperacetylation of tubulin disrupts autophagosome Des 84.28 7.79 7.93 traffic, we performed live imaging of EGFP-LC3 trafficking 15 A2E* 91.43 5.28 3.29 in RPE treated with trichostatin A (TSA), an inhibitor of A2E + des 83.88 7.77 8.35 histone deacetylase 6 (HDAC6), the enzyme that deacety- -*,-ig-ni-fic-an-tl_y_d_iffi-er-en-t-fro_m_a_ll_o_th_er_c_on-d1-·tio-n-,,-p-<-O-.O-OO-l,-o-ne--w-a-y-A-N-O-VA-.-- lates tubulin (Joseph et al., 2008). Confirming our hypoth- esis, TSA treatment replicated the autophagosome traffick ing defects seen in cells with A2E (FIGS. 6F, 6G; Table 2). 20 Discussion Thus, bisretinoid-induced cholesterol storage prevents autophagosome trafficking by increasing tubulin acetylation. Here, we describe a novel molecular mechanism by which autophagy is derailed by lipofuscin bisretinoids and excess Acid Sphingomyelinase Activation Promotes cholesterol (FIG. 10), which progressively accumulate in the Hyperacetylation of Tubulin in the RPE 25 RPE and contribute to the pathogenesis of macular degen erations (Ambati et al., 2013; Bowes Rickman et al., 2013; To dissect the molecular mechanism that links bisretinoid Pikuleva and Curcio, 2014). Although decreased autophagy stimulated cholesterol accumulation in the endo-lysosomal within the retina is thought to participate in the pathogenesis system with tubulin acetylation, we sought clues from ofretinal dystrophies (Bowes Rickman et al., 2013; Frost et cholesterol-storage disorders like Niemann-Pick Cl 30 al., 2014), little is currently known about the precise mecha (NPCl). In NPCl fibroblasts, excess cholesterol in late nisms involved or how autophagy can be exploited as a endosomes and lysosomes sequesters the anionic phospho potential drug target to maintain RPE health. lipid BMP (Pipalia et al., 2007). BMP is a co-factor for acid The data presented in this study identify ASMase as a sphingomyelinase (ASMase), the lysosomal enzyme that critical regulator of autophagy in RPE compromised by hydrolyzes sphingomyelin to generate ceramide (Kirkegaard 35 lipofuscin-mediated cholesterol accumulation. We show that et al., 2010). Recent studies show that ceramide regulates in cells with bisretinoids, cholesterol sequesters the anionic tubulin acetylation via atypical protein kinase C (aPKC) and lipid BMP within RPE late endosomes and lysosomes. BMP aurora A kinase (He et al., 2012; He et al., 2014). In activates ASMase, the enzyme that hydrolyzes sphingomy polarized primary RPE with A2E, immunofluorescence elin to ceramide, which in tum promotes tubulin acetylation imaging showed high levels of BMP, which co-localized 40 on stable microtubules. Studies in polarized epithelia and with filipin staining for cholesterol (FIG. 7A). Cells with neural progenitors show that ceramide inhibits the microtu A2E also had high ASMase activity (FIG. 7B) and more bule deacetylase HDAC6 by preventing the translocation of ceramide (FIG. 7C) compared to control RPE. Treatment aPKC from the membrane to the cytosol. This interferes with desipramine, a functional inhibitor of ASMase (Kom with aPKC-mediated phosphorylation of two HDAC6 acti- huber et al., 2010), decreased AS Mase activity and ceramide 45 vators, aurora A kinase and glycogen synthase kinase 3~, levels (FIGS. 7B, 7C). Desipramine also decreased acety resulting in the accumulation of acetylated tubulin (He et al., lated tubulin in cells with A2E (FIG. 7D), confirming that 2012; He et al., 2014). Ceramide can also phosphorylate the cholesterol increases tubulin acetylation via a BMP-AS focal adhesion scaffold protein paxillin (Sasaki et al., 1996), Mase-ceramide pathway. which has been recently identified as a negative regulator of 50 HDAC6 activity (Deakin and Turner, 2014). Whether any of Inhibition of ASMase Activity Corrects Autophagic these mechanisms are responsible for ceramide-induced Defects in the RPE tubulin acetylation in RPE with bisretinoids remains to be determined. If, as the above data indicate, ASMase is a critical Live imaging data showed impaired autophagosome bio- regulator of autophagy, then ASMase inhibition should be 55 genesis and trafficking as a consequence of increased tubulin sufficient to restore autophagic flux in RPE with bisretinoid acetylation, either due to lipofuscin bisretinoids or after mediated cholesterol accumulation. We first performed live treatment with the HDAC6 inhibitor TSA. Acetylation is a imaging of EFGP-LC3 and image analyses of trafficking post-translational modification of a-tubulin that can act data (FIG. SA) showed that desipramine increased both the either singly or in concert with other modifications such as number of motile tracks and long-range displacement of 60 tyrosination/detyrosination to control motor recruitment in autophagosomes in RPE cells withA2E (FIGS. SB-E, Table cargo-specific manner (Hammond et al., 2008; Mackeh et 3). After mTOR inhibition, a short exposure to desipramine al., 2013). Precisely how post-translational modifications of increased LC3B-II levels in RPE with A2E comparable to tubulin modulate the trafficking of autophagosomes and those in control cells (FIGS. 9A, 9B). Desipramine also other organelles is not well understood. Acetylated micro- decreased p62 levels in cells with A2E (FIG. 9A, 9C), 65 tubules in neurons preferentially recruit kinesin-1 and the indicating a restoration of autophagic flux. In agreement scaffolding protein JIPl (JNK-interacting protein 1) to direct with immunoblotting data, tfLC3 imaging showed that polarized traffic to a subset of neurites (Reed et al., 2006). US 10,124,008 B2 19 20 Binding of JIPl to the kinesin heavy chain (KHC) motor late that ASMase inhibition could be a novel therapeutic domain ofkinesin-1 accelerates anterograde traffic, whereas approach not only for retinal degenerations, but also for JIPl binding to the pl50GZued subunit of the dynein-dynactin neurodegenerative diseases. complex promotes retrograde traffic. JIPl has been recently shown to bind LC3 in neurons to direct dynein-mediated 5 Example 2. ASMase Inhibition Restores retrograde transport of autophagosomes. Interestingly, the Cell-Surface Complement Regulatory Protein LC3-JIP1 interaction interferes with JIPl-mediated activa Levels in RPE with Bisretinoids tion of kinesin-1 (Fu et al., 2014). Unlike autophagosomes in neurons, which undergo unidirectional retrograde trans The alternative pathway of the complement system has port along the axon (Fu et al., 2014), our data show that 10 been implicated in the pathogenesis of AMD. The final step autophagosomes in the RPE exhibit bidirectional motility, of the complement pathway is the formation of the mem likely driven by opposing actions of kinesin and dynein brane attack complex (MAC), which forms pores in cell motors (Fu and Holzbaur, 2014). It is possible that acety membranes and can cause cell lysis. Lipofuscin and A2E have been shown to activate complement components in lated microtubules in RPE with bisretinoids preferentially 15 recruit kinesin-1 (Reed et al., 2006), which would then mouse models of Stargardt disease and in cultured RPE compete with LC3 for JIPl binding (Fu et al., 2014). We also (Radu et al., 2014; Zhou et al., 2009). The RPE have observed fewer tyrosinated microtubules in cells with A2E, numerous mechanisms that protect against complement which could interfere with the recruitment of pl50GZued; mediated damage such as (a) cell-surface complement regu dynactin (Rocha et al., 2009). Thus, in RPE with bis- 20 latory proteins like CD55, CD59 and CD46, which prevent retinoids, increased acetylation and decreased tyrosination specific steps of MAC assembly and (b) resealing MAC oftubulin could interfere with bidirectional autophagosome pores by lysosome exocytosis. transport, possibly by altering the recruitment of motor CD55 and CD59 are GPI-anchored proteins whose deliv proteins and/or preventing interactions between motors, ery to the plasma membrane depends on lysosomal choles scaffolds and cargo. Further studies will help dissect the 25 terol (Mayor et al., 2004). Cholesterol depletion speeds the roles of these motor and scaffolding proteins in directing the transport of CD55 and CD59 to the cell surface by recycling transport of autophagosomes and other organelles (endo endosomes, whereas cholesterol overload shuttles these somes, lysosomes, phagosomes, etc.) in the RPE. GPI-anchored proteins towards lysosomal degradation How might increased tubulin acetylation interfere with (Mayor et al., 1998). Since our data show that bisretinoids autophagosome biogenesis? At the earliest stages of 30 interfere with long-range microtubule-mediated organelle autophagosome formation, phosphatidylinositol 3-phos transport in the RPE, we first examined whether delivery of phate binds its effectors WIPil and WIPI2 to catalyze the CD59 is affected in RPE with bisretinoids. Because CD59 sequential recruitment of Atg proteins that regulate elonga prevents MAC formation, we also examined the efficiency tion of the preautophagosomal membrane. The fully formed of membrane repair by lysosome exocytosis in the RPE after 35 complement attack. Finally, we measured the levels of Atg5-Atgl2-Atg16L complex induces covalent conjugation reactive oxygen species (ROS) in RPE after complement of phosphatidylethanolamine to LC3 and facilitates attack because sustained influx of calcium through MAC autophagosome closure. Movement of these preautophago pores can cause mitochondrial damage and increase ROS somes along dynamic microtubules is necessary for bothAtg generation. recruitment and for driving subsequent steps of autophago- 40 some formation (Geeraert et al., 2010). It is therefore likely Materials and Methods that increased stability of acetylated microtubules in cells with A2E interferes with the recruitment of Atg proteins, Immunostaining which decreases membrane elongation and autophagosome biogenesis. 45 Filter-grown primary polarized RPE were fixed in 2% Our data demonstrate that desipramine, a tricyclic anti paraformaldehyde (EMS) for 10 min, blocked in 1% BSA in 2 2 depressant that increases AS Mase proteolysis (Kornhuber et PBS supplemented with Ca + and Mg + and incubated with al., 2010), restores autophagy in RPE withA2E by reversing specific primary antibodies for 1 hour: mouse anti-CD59 ceramide-induced tubulin acetylation. Thus, functional (clone MEM43, 1:200, Pierce MAl-19133), rabbit anti ASMase inhibitors, many of which are FDA-approved drugs 50 Lamp2a (1:200, Abeam ab18528), goat anti-EEAl (1:500, with established safety and efficacy profiles (Kornhuber et Santa Cruz sc-6415), mouse anti-C5b-9 (1:100, Novus al., 2010), are promising candidates for inherited macular NBPl-05120), rabbit anti-C9b/C9 (1:200, Bioss dystrophies characterized by elevated levels of lipofuscin BS-15307R), and rat anti-ZO-1 (1:3000)(23). AlexaFluor bisretinoids such as Stargardt and Best diseases (Travis et secondary antibodies were used at 1 :500. Filters were al., 2007). Lipofuscin bisretinoids are also implicated in 55 mounted under coverslips on glass slides under AMD, the most common cause of vision loss in older adults VECTASHIELD (Vector labs), sealed and visualized with (Ambati et al., 2013). In this context, it is intriguing to note Andor Revolution XD spinning disk confocal microscope that allelic variants in cholesterol transporters and lipopro using 60xl .4 NA or l00xl .49 NA oil objectives with tein metabolizing enzymes modulate susceptibility to AMD identical exposures and gains for each antibody. Quantifi (Fritsche et al., 2014) and that use of tricyclic antidepres- 60 cation of surface CD59 staining was performed using sants like desipramine is associated with a statistically Imaris. significant decrease in the risk of developing early AMD (van Leeuwen et al., 2004). Furthermore, ASMase activity Lysosome Exocytosis and ceramide levels are increased in the brains of patients with Alzheimer's and Parkinson's diseases (Haughey et al., 65 Cells were rinsed in recording medium (HBSS with 4.5 2010; Fabela et al., 2011), which are associated with dys g/L glucose, 20 mM HEPES) and incubated with ionomycin regulated autophagy (Nixon, 2013). It is tempting to specu- or 1a% normal human serum (Quidel) for 10 min at 37° C. US 10,124,008 B2 21 22 Filters were immediately transferred to ice and analyzed for membrane repair resulted in increased production of ROS in surface LAMP2 or ~-hex activity (see below). RPE with A2E (FIG. llD). A short treatment with desipra mine increased surface CD59 levels, restored lysosome Detection of Cell-Surface Lamp2 and C5b-9 exocytosis and decreased ROS levels after complement 5 attack (FIGS. llA, llC & D). The RPE are antigen Cells were stained with mouse monoclonal antibody to presenting cells and expression of major histocompatibility the lumenal domain of Lamp2 (Serotec, MCA2558, 1:500) complex II (MHC-II) in the RPE and outer retina is (48) or C5b-9 (Novus, NBPl-05120, 1:100) at 4° C. for 30 increased in inflammatory conditions and in AMD (Penfold min. The antibody was diluted in PBS supplemented with et al., 1997). Desipramine decreased MHC-II protein levels 2 2 10 Ca + and Mg + + 1% BSA. Cells were then fixed with 2% in the RPE with A2E (FIG. llE). PFA for 5 min on ice, permeabilized, and stained for ZO-1 as detailed above. Alexa-conjugated secondary antibodies Conclusion (INVITROGEN) were used at 1 :500 for 30 min. Cells were imaged by confocal microscopy (Andor Revolution XD) Thus, lipofuscin bisretinoids appear to promote a pro using a 60x 1 .4 NA oil objective. For each set of experi- 15 inflammatory environment in the RPE and retina in vivo and ments, the laser power, voltage and offset were identical for our data support the hypothesis that ASMase inhibition is a a given fluorophore. Quantification of Lamp2 staining was valid strategy to decrease inflammation in the RPE. performed by obtaining the total fluorescence using the surface tool in Imaris (Bitplane). Example 3. Genetic Validation of ASMase as a 20 Therapeutic Target in RPE with Bisretinoids Measurement of ~-Hex Activity To establish that inhibition of ASMase does indeed After drug treatments, apical and basolateral media were decrease tubulin acetylation and increase autophagy in RPE collected, centrifuged at l00xg for 5 min to pellet dead cells with bisretinoids, we used shRNA-mediated knockdown of and 10,000xg for 5 min to pellet debris. Cells were lysed in 25 ASMase. 0.5 ml PBS+ 1% NP-40 for total ~-hex activity. To measure Materials and Methods: Primary porcine RPE were trans enzyme activity, 350 µI of supernatant was incubated for 20 fected with plasmids expressing RFP-tagged shRNA con min with 50 µI of 6 mM 4-methyl-umbelliferyl-N-acetyl-~ structs (Origene) to porcine ASMase (SMPDl gene) and D-glucosaminide (Sigma) in sodium citrate-phosphate buf exposed to A2E 48 h after transfection as described previ fer, pH 4.5. Fluorescence was measured after stopping the 30 ously. Cells were fixed 48 h after A2E loading and stained reaction with 100 µI 2 M Na2CO3, 1.1 M glycine (365 nm for acetylated tubulin or LC3 to label autophagosomes. excitation, 450 nm emission, Tecan microplate reader). Cell Protein knockdown was confirmed by immunoblotting for extracts were diluted 1:50 before assay for total cellular ASMase. ~-hex activity. Results: We tested 4 different shRNA constructs and 35 found that construct D elicited the most robust knockdown Reactive Oxygen Species Assay of ASMase in porcine RPE (FIG. 12A). Similar to our data with desipramine, depletion of ASMase decreased acety Reactive oxygen species (ROS) generation was measured lated tubulin in RPE with bisretinoids (FIG. 12B). Compare using CELLROX Deep Red (Molecular Probes, #10422) immunostaining for ASMase (green) and acetylated tubulin according to manufacturer's recommendation. After 10% 40 (white) in shRNA-expressing cells (RFP-positive, yellow NHS treatment, medium containing NHS was removed and asterisk) with those not transfected (white #). ASMase cells were rinsed with HBSS before incubation in 5 µM depletion increased autophagosome biogenesis in RPE with CELLROX reagent at 37° C. for 30 min. Cells were rinsed A2E to control levels whereas the scrambled shRNA 3 times with PBS and fluorescence was read with microplate sequence had no effect (FIG. 12C). reader at Ex/Em of 640/665. For normalization purposes, 45 cells were fixed, stained with DAPI, and fluorescence was Example 4. Evaluation of ASMase Inhibitors in an read at Ex/Em: 360/460. in Vitro Model of RPE Dysfunction Results ASMase is a soluble, zinc-dependent enzyme that binds 50 the lumenal surface of lysosomal membranes to hydrolyze We observed strikingly less CD55 and CD59 on the sphingomyelin to ceramide. Rational design of potent, selec 1 plasma membrane in RPE from the Abca4_ _ mouse model tive ASMase inhibitors is challenging because the crystal of Stargardt disease (FIG. llA) and in cells with A2E (FIG. structure of the enzyme has not yet been resolved. Desip llB), suggesting that bisretinoid-induced cholesterol stor ramine and other cationic amine drugs act as functional age likely prevents the recycling these GPI-anchored 55 ASMase inhibitors: they inhibit binding of ASMase to the complement-regulatory proteins. Since CD59 inhibits the anionic lipid BMP, which is required for ASMase activity formation of the membrane attack complex, this would (Kirkegaard, et al., 2010). When AS Mase can no longer bind render the RPE susceptible to complement-mediated dam BMP, it detaches from the lysosomal membrane and age. Lysosome exocytosis reseals pores formed by C5b-9 becomes susceptible to lysosomal proteolysis (Kaizer, et al., and prevents cell lysis (Xu et al., 2012). In control RPE 60 2004 ). Therefore, the efficacy of functional AS Mase inhibi monolayers, exposure to 10% normal human serum (NHS) tors depends on their ability to accumulate in lysosomes. as a source of complement induced lysosome exocytosis All functional ASMase inhibitors identified thus far are monitored by appearance of the lysosome membrane protein low molecular weight ( <500 Da) lipophilic weak bases, with LAMP2 on the cell surface (FIG. llC, green-LAMP2, at least one basic nitrogen atom. Although pKa and lipophi- red-tight junction protein ZO-1 to demarcate cell bound 65 licity are important, not all lipophilic weak bases are aries). Lysosome exocytosis was severely blunted in cells ASMase inhibitors; structure-activity relationship models with A2E after complement exposure. This inhibition of show that the steric hindrance of the protonated nitrogen is US 10,124,008 B2 23 24 more important than pKa because it shields the enzyme from effective as desipramine in decreasing acetylated tubulin in 42 binding BMP . Several bisphosphonates such as zoledronic cells with bisretinoids (FIG. 13A). At an equivalent dose, acid, currently used to treat osteoporosis, inhibit ASMase zoledronic acid appears to be more effective than the func 2 structurally by forming active complexes with the Zn + ion tional inhibitors, suggesting that structural ASMase inhibi- 62 in the active center of the enzyme . It is thought that 5 tors could improve RPE function at lower doses than func structural inhibitors can inhibit ASMase at lower concen tional inhibitors. Fexofenadine is not as effective as trations than functional inhibitors. astemizole, sertraline or desipramine in increasing Here, we tested the abilities of functional and structural autophagic flux in the RPE. Although astemizole is an ASMase inhibitors (Table 4) to decrease acetylated tubulin effective AS Mase inhibitor, it is no longer marketed because and induce autophagy in our well-characterized polarized 10 of rare cardiovascular side-effects. Our data indicate that primary adult RPE model (Toops, et al. 2014) in the pres astemizole and fexofenadine can be excluded from this list. ence of lipofuscin bisretinoids. Further, clomiphene, maprotiline and promethazine will also Materials and Methods likely be excluded due to serious side effects. Based on these 15 results, five to seven additional ASMase inhibitors may We performed two initial screens: first, we compared the prove to be as effective as desipramine in increasing ability of functional inhibitors (fluoxetine and amitriptyline) autophagy and decreasing inflanimation in the RPE. Since with that of the lone structural AS Mase inhibitor zoledronic zoledronic acid acts by a different mechanism than func acid in decreasing acetylated tubulin. Here, polarized RPE tional inhibitors and dramatically decreases acetylated tubu- monolayers with or without A2E were treated with one of 20 !in in RPE with bisretinoids, combination therapy with the three inhibitors (10 µM, 3 h). Desipramine as a positive structural and functional inhibitors could prove to be very control. At the end of the incubation period, cells were fixed efficacious in maintaining RPE health and preventing vision and stained for acetylated tubulin, imaged and the amount of loss. acetylated tubulin in the cells was quantified. Second, we evaluated three additional drugs (sertraline, astemizole and 25 REFERENCES fexofenadine) for their abilities to increase autophagy in the RPE. Cells with or without bisretinoids were transfected Ambati, J., Atkinson, J. P., and Gelfand, B. D. (2013). with mRFP-GFP-LC3, starved to induce autophagy, treated Immunology of age-related macular degeneration. Nat with drugs and imaged as in FIG. 9D. Rev Immunol 13, 438-451. TABLE 4 FDA-approved ASMase inhibitors. Bulk Drug* Current indication supplier Potential issues Desiprarnine Tricyclic antidepressant (TCA); TOCRIS tertiary amine Arnitriptyline TCA (tertiary amine) Selleck Chem Arnlodipine Calcium channel blocker TOCRIS Astemizole H2-receptor antagonist TOCRIS Benzatropine Anti-cholinergic SIGMA- ALDRICH Citalopram Selective serotonin reuptake TOCRIS Can cause inhibitor (SSRI) arrhythmia at high doses. Clomiphene Selective estrogen receptor Selleck May have serious modulator Chem side effects. Cloperastine Cough suppressant (anti- SIGMA- histamine) ALDRICH Cyclobenzaprine Skeletal muscle relaxant SIGMA- ALDRICH Cyproheptadine Histamine Hl antagonist TOCRIS Diphenhydramine Histamine Hl antagonist TOCRIS Fexofenadine Histamine Hl antagonist TOCRIS Fluoxetine SSRI TOCRIS Maprotiline Tetracyclic antidepressant TOCRIS Contraindicated in glaucoma Nortriptyline TCA (secondary amine) SIGMA- ALDRICH Paroxetine SSRI TOCRIS Promethazine Histamine Hl antagonist SIGMA- Contraindicated in ALDRICH glaucoma Sertraline Selective serotonin reuptake TOCRIS inhibitor Zoledronic acid Bisphosphonate to treat Enzo <4% oral osteoporosis Biochem bioavailability * All drugs will be used at 0-10 µM for in vitro studies; minimum effective dose established in vitro will be used for ex vivo studies; oral doses may be based on literature reports (e.g., Gulbins et al., 2013). 65 The effectiveness ofAS Mase inhibitors is seen in FIG. 13. Ambati, J., and Fowler, B. J. (2012). Mechanisms of age Our results indicate that fluoxetine and amitriptyline are as related macular degeneration. Neuron 75, 26-39. US 10,124,008 B2 25 26 Barmada, S. J., Serio, A., Arjun, A., Bilican, B., Daub, A., Codogno, P., and Pous, C. (2010). Starvation-induced Ando, D. M., Tsvetkov, A., Pleiss, M., Li, X., Peisach, D., hyperacetylation oftubulin is required for the stimulation Shaw, C., Chandran, S., and Finkbeiner, S. (2014). of autophagy by nutrient deprivation. J Biol Chem 285, Autophagy induction enhances TDP43 turnover and sur 24184-24194. vival in neuronal ALS models. Nat Chem Biol 10, 677- 5 Grumati, P., Coletta, L., Sabatelli, P., Cescon, M., Angelin, 685. A., Bertaggia, E., Blaauw, B., Urciuolo, A., Tiepolo, T., Besirli, C. G., Chinskey, N. D., Zheng, Q. D., and Zacks, D. Merlini, L., Maraldi, N. M., Bernardi, P., Sandri, M., and N. (2011). Autophagy activation in the injured photore Bonaldo, P. (2010). Autophagy is defective in collagen VI ceptor inhibits fas-mediated apoptosis. Invest Ophthalmol muscular dystrophies, and its reactivation rescues myo Vis Sci 52, 4193-4199. 10 fiber degeneration. Nat Med 16, 1313-1320. Bok, D. (1993). The retinal pigment epithelium: a versatile Gulbins, E. et al. Acid sphingomyelinase-ceramide system partner in vision. J Cell Sci Suppl 17, 189-195. mediates effects of antidepressant drugs. Nat Med 19, Bowes Rickman, C., Farsiu, S., Toth, C. A., and Klingeborn, 934-938 (2013). M. (2013). Dry age-related macular degeneration: mecha Haji Abdollahi, S., Hirose, T., 2013. Stargardt-Fundus fla nisms, therapeutic targets, and imaging. Invest Ophthal- 15 vimaculatus: recent advancements and treatment. Semin mo! Vis Sci 54, ORSF68-80. Ophthalmo128, 372-376. Chen, Y., Sawada, 0., Kohno, H., Le, Y. Z., Subauste, C., Hammond, J. W., Cai, D., and Verhey, K. J. (2008). Tubulin Maeda, T., and Maeda, A. (2013). Autophagy protects the modifications and their cellular functions. Curr Opin Cell retina from light-induced degeneration. J Biol Chem 288, Biol 20, 71-76. 7506-7518. 20 Haughey, N. J., Bandaru, V. V., Bae, M., and Mattson, M. P. Choi, A. M., Ryter, S. W., and Levine, B. (2013). Autophagy (2010). Roles for dysfunctional sphingolipid metabolism in human health and disease. N Engl J Med 368, 651-662. in Alzheimer's disease neuropathogenesis. Biochim Bio Codogno, P., Mehrpour, M., and Proikas-Cezanne, T. phys Acta 1801, 878-886. (2012). Canonical and non-canonical autophagy: varia He, Q., Wang, G., Dasgupta, S., Dinkins, M., Zhu, G., and tions on a common theme of self-eating? Nat Rev Mo! 25 Bieberich, E. (2012). Characterization of an apical cer Cell Biol 13, 7-12. amide-enriched compartment regulating ciliogenesis. Mo! Deakin, N. 0., and Turner, C. E. (2014). Paxillin inhibits Biol Cell 23, 3156-3166. HDAC6 to regulate microtubule acetylation, Golgi struc He, Q., Wang, G., Wakade, S., Dasgupta, S., Dinkins, M., ture, and polarized migration. J Cell Biol 206, 395-413. Kong, J. N., Spassieva, S. D., and Bieberich, E. (2014). Doyle, S. L., Ozaki, E., Brennan, K., Humphries, M. M., 30 Primary cilia in stem cells and neural progenitors are Mulfaul, K., Keaney, J., Kenna, P. F., Maminishkis, A., regulated by neutral sphingomyelinase 2 and ceramide. Kiang, A. S., Saunders, S. P., Hams, E., Lavelle, E. C., Mo! Biol Cell 25, 1715-1729. Gardiner, C., Fallon, P. G., Adamson, P., Humphries, P., Jimenez-Sanchez, M., Menzies, F. M., Chang, Y. Y., and Campbell, M. (2014). IL-18 attenuates experimental Simecek, N., Neufeld, T. P., and Rubinsztein, D. C. choroidal neovascularization as a potential therapy for 35 (2012). The Hedgehog signalling pathway regulates wet age-related macular degeneration. Sci Transl Med 6, autophagy. Nat Commun 3, 1200. 230ra244. Joseph, R. A., Shepard, B. D., Kannarkat, G. T., Rutledge, T. Eldred, G. E., and Lasky, M. R. (1993). Retinal age pigments M., Tuma, D. J., and Tuma, P. L. (2008). Microtubule generated by self-assembling lysosomotropic detergents. acetylation and stability may explain alcohol-induced Nature 361, 724-726. 40 alterations in hepatic protein trafficking Hepatology 47, Fabela, N., Martin, V., Santpere, G., Marin, R., Torrent, L., 1745-1753. Ferrer, I., and Diaz, M. (2011). Severe alterations in lipid Kim, J. Y., Zhao, H., Martinez, J., Doggett, T. A., Kole composition of frontal cortex lipid rafts from Parkinson's snikov, A. V., Tang, P. H., Ablonczy, Z., Chan, C. C., disease and incidental Parkinson's disease. Molecular Zhou, Z., Green, D. R., and Ferguson, T. A. (2013). medicine 17, 1107-1118. 45 Noncanonical autophagy promotes the visual cycle. Cell Fraldi, A., Annunziata, F., Lombardi, A., Kaiser, H. J., 154, 365-376. Medina, D. L., Spampanato, C., Fedele, A. 0., Polish Kirkegaard, T., Roth, A. G., Petersen, N. H., Mahalka, A. K., chuk, R., Sorrentino, N. C., Simons, K., and Ballabio, A. Olsen, 0. D., Moilanen, I., Zylicz, A., Knudsen, J., (2010). Lysosomal fusion and SNARE function are Sandhoff, K., Arenz, C., Kinnunen, P. K., Nylandsted, J., impaired by cholesterol accumulation in lysosomal star- 50 and Jaattela, M. (2010). Hsp70 stabilizes lysosomes and age disorders. EMBO J 29, 3607-3620. reverts Niemann-Pick disease-associated lysosomal Fritsche, L. G., Fariss, R. N., Stambolian, D., Abecasis, G. pathology. Nature 463, 549-553. R., Curcio, C. A., and Swaroop, A. (2014). Age-Related Klein, R., Klein, B. E., Jensen, S. C., Cruickshanks, K. J., Macular Degeneration: Genetics and Biology Coming Lee, K. E., Danforth, L. G., Tomany, S. C., 2001. Medi Together Annu Rev Genomics Hum Genet. 55 cation use and the 5-year incidence of early age-related Frost, L. S., Mitchell, C. H., and Boesze-Battaglia, K. maculopathy: the Beaver Dam Eye Study. Arch Ophthal- (2014). Autophagy in the eye: Implications for ocular cell mol 119, 1354-1359. health. Exp Eye Res. Klionsky, D. J., et al. (2012). Guidelines for the use and Fu, M. M., and Holzbaur, E. L. (2014). Integrated regulation interpretation of assays for monitoring autophagy. of motor-driven organelle transport by scaffolding pro- 60 Autophagy 8, 445-544. teins. Trends Cell Biol. Ko, D. C., Gordon, M. D., Jin, J. Y., and Scott, M. P. (2001). Fu, M. M., Nirschl, J. J., and Holzbaur, E. L. (2014). LC3 Dynamic movements of organelles containing Niemann binding to the scaffolding protein JIPl regulates proces Pick Cl protein: NPCl involvement in late endocytic sive dynein-driven transport of autophagosomes. Dev events. Mo! Biol Cell 12, 601-614. Cell 29, 577-590. 65 Koga, H., Kaushik, S., and Cuervo, A. M. (2010). Altered Geeraert, C., Ratier, A., Pfisterer, S. G., Perdiz, D., Canta lipid content inhibits autophagic vesicular fusion. FASEB loube, I., Rouault, A., Pattingre, S., Proikas-Cezanne, T., J 24, 3052-3065. US 10,124,008 B2 27 28 Kaizer, M., Werth, N., Sandhoff, K., 2004. Interactions of Pampliega, 0, Orhon, I., Patel, B., Sridhar, S., Diaz-Carret acid sphingomyelinase and lipid bilayers in the presence ero, A., Beau, I., Codogno, P., Satir, B. H., Satir, P., and of the tricyclic antidepressant desipramine. FEBS Lett Cuervo, A. M. (2013). Functional interaction between 559, 96-98. autophagy and ciliogenesis. Nature 502, 194-200. Kornhuber, J., Tripal, P., Reichel, M., Muhle, C., Rhein, C., 5 Perdiz, D., Mackeh, R., Pous, C., and Baillet, A. (2011 ). The Muehlbacher, M., Groemer, T. W., andGulbins, E. (2010). ins and outs of tubulin acetylation: more than just a Functional Inhibitors of Acid Sphingomyelinase (FIAS post-translational modification? Cell Signal 23, 763-771. MAs ): a novel pharmacological group of drugs with broad Pikuleva, I. A., and Curcio, C. A. (2014). Cholesterol in the clinical applications. Cellular physiology and biochemis retina: The best is yet to come. Prag Retin Eye Res. try: international journal of experimental cellular physi- 10 Pipalia, N. H., Hao, M., Mukherjee, S., and Maxfield, F. R. ology, biochemistry, and pharmacology 26, 9-20. (2007). Stern!, protein and lipid trafficking in Chinese Kreitzer, G., Schmoranzer, J., Low, S. H., Li, X., Gan, Y., hamster ovary cells with Niemarm-Pick type Cl defect. Weimbs, T., Simon, S. M., and Rodriguez-Boulan, E. (2003). Three-dimensional analysis of post-Golgi carrier Traffic 8, 130-141. Polson, H. E., de Lartigue, J., Rigden, D. J., Reedijk, M., exocytosis in epithelial cells. Nat Cell Biol 5, 126-136. 15 Kubota, R., Boman, N. L., David, R., Mallikaarjun, S., Patil, Urbe, S., Clague, M. J., and Tooze, S. A. (2010). Mam S., Birch, D., 2012. Safety and effect on rod function of malian Atgl8 (WIPI2) localizes to omegasome-anchored ACU-4429, a novel small-molecule visual cycle modula phagophores and positively regulates LC3 lipidation. tor. Retina 32, 183-188. Autophagy 6, 506-522. Kunchithapautham, K., and Rohrer, B. (2007a). Apoptosis 20 Radu, R. A., Hu, J., Yuan, Q., Welch, D. L., Makshanoff, J., and autophagy in photoreceptors exposed to oxidative Lloyd, M., McMullen, S., Travis, G. H., and Bok, D. stress. Autophagy 3, 433-441. (2011 ). Complement system dysregulation and inflamma Kunchithapautham, K., and Rohrer, B. (2007b ). Autophagy tion in the retinal pigment epithelium of a mouse model is one of the multiple mechanisms active in photoreceptor for Stargardt macular degeneration. J Biol Chem 286, degeneration. Autophagy 3, 65-66. 25 18593-18601. Lakkaraju, A., Finnemann, S. C., and Rodriguez-Boulan, E. Radu R A, Hu J, Jiang Z, & Bok D (2014) Bisretinoid (2007). The lipofuscin fluorophore A2E perturbs choles mediated complement activation on retinal pigment epi terol metabolism in retinal pigment epithelial cells. Proc thelial cells is dependent on complement factor H haplo Natl Acad Sci USA 104, 11026-11031. type. J Biol Chem 289 (13):9113-9120. Le Guezennec, X., Brichkina, A., Huang, Y. F., Kostromina, 30 Reed, N. A., Cai, D., Blasius, T. L., Jih, G. T., Meyhofer, E., E., Han, W., and Bulavin, D. V. (2012). Wipl-dependent Gaertig, J., and Verhey, K. J. (2006). Microtubule acety regulation of autophagy, obesity, and atherosclerosis. Cell lation promotes kinesin-1 binding and transport. Current Metab 16, 68-80. biology: CB 16, 2166-2172. Lebrand, C., Corti, M., Goodson, H., Casson, P., Cavalli, V., Reme, C. E., Wolfrum, U., Imsand, C., Hafezi, F., and Mayran, N., Faure, J., and Gruenberg, J. (2002). Late 35 Williams, T. P. (1999). Photoreceptor autophagy: effects endosome motility depends on lipids via the small of light history on number and opsin content of degrada GTPase Rab7. EMBO J 21, 1289-1300. tive vacuoles. Invest Ophthalmol Vis Sci 40, 2398-2404. Lee, J. K., Jin, H. K., Park, M. H., Kim, B. R., Lee, P. H., Rocha, N., Kuijl, C., van der Kant, R., Janssen, L., Houben, Nakauchi, H., Carter, J. E., He, X., Schuchman, E. H., and D., Janssen, H., Zwart, W., and Neefjes, J. (2009). Cho Bae, J. S. (2014). Acid sphingomyelinase modulates the 40 lesterol sensor ORPlL contacts the ER protein YAP to autophagic process by controlling lysosomal biogenesis control Rab7-RILP-p150 Glued and late endosome posi- in Alzheimer's disease. J Exp Med 211, 1551-1570. tioning. J Cell Biol 185, 1209-1225. Liu, D., Meckel, T., and Long, E. 0. (2010). Distinct role of Rodriguez-Boulan, E., Kreitzer, G., and Musch, A. (2005). rab27a in granule movement at the plasma membrane and Organization of vesicular trafficking in epithelia. Nat Rev in the cytosol of NK cells. PLoS One 5, el 2870. 45 Mo! Cell Biol 6, 233-247. Lois, N., McBain, V., Abdelkader, E., Scott, N. W., Kumari, Rodriguez-Muela, N., Koga, H., Garcia-Ledo, L., de la R., 2013. Retinal pigment epithelial atrophy in patients Villa, P., de la Rosa, E. J., Cuervo, A. M., and Boya, P. with exudative age-related macular degeneration under (2013). Balance between autophagic pathways preserves going anti-vascular endothelial growth factor therapy. retinal homeostasis. Aging Cell 12, 478-488. Retina 33, 13-22. 50 Roth, A. G., Drescher, D., Yang, Y., Redmer, S., Uhlig, S., Mackeh, R., Perdiz, D., Lorin, S., Codogno, P., and Pous, C. Arenz, C., 2009. Potent and selective inhibition of acid (2013). Autophagy and microtubules-new story, old sphingomyelinase by bisphosphonates. Angew Chem Int players. Journal of cell science 126, 1071-1080. Ed Engl 48, 7560-7563. Mata, N. L., J. Weng, and G. H. Travis. 2000. Biosynthesis Rubinsztein, D. C., Gestwicki, J. E., Murphy, L. 0., and of a major lipofuscin fluorophore in mice and humans 55 Klionsky, D. J. (2007). Potential therapeutic applications with ABCR-mediated retinal and macular degeneration. of autophagy. Nat Rev Drug Discov 6, 304-312. Proc Natl Acad Sci USA. 97:7154-7159. Sarkar, S., Carroll, B., Buganim, Y., Maetzel, D., Ng, A. H., Mayor, S. & Riezman, H. Sorting GPI-anchored proteins. Cassady, J.P., Cohen, M.A., Chakraborty, S., Wang, H., Nat Rev Mal Cell Biol 5, 110-120 (2004). Spooner, E., Ploegh, H., Gsponer, J., Korolchuk, V. I., and Mayor, S., Sabharanjak, S. & Maxfield, F. R. Cholesterol- 60 Jaenisch, R. (2013). Impaired autophagy in the lipid dependent retention of GPI-anchored proteins in endo storage disorder Niemann-Pick type Cl disease. Cell somes. EMBO J 17, 4626-4638. (1998). reports 5, 1302-1315. Meleth, A. D., Wong, W. T., Chew, E. Y., 2011. Treatment for Sasaki, T., Hazeki, K., Hazeki, 0, Ui, M., and Katada, T. atrophic macular degeneration. Curr Opin Ophthalmol 22, (1996). Focal adhesion kinase (p125FAK) and paxillin are 190-193. 65 substrates for sphingomyelinase-induced tyrosine phos Nixon, R. A. (2013). The role of autophagy in neurodegen phorylation in Swiss 3T3 fibroblasts. Biochem J 315 (Pt erative disease. Nat Med 19, 983-997. 3), 1035-1040. US 10,124,008 B2 29 30 Sparrow, J. R., Gregory-Roberts, E., Yamamoto, K., Zhou J, Kim S R, Westlund B S, & Sparrow J R (2009) Blonska, A., Ghosh, S. K., Ueda, K., and Zhou, J. (2012). Complement activation by bisretinoid constituents of The bisretinoids ofretinal pigment epithelium. Prag Retin RPE lipofuscin. Invest Ophthalmol Vis Sci 50 (3 ): 1392- Eye Res 31, 121-135. 1399. Sparrow, J. R., C. A. Parish, M. Hashimoto, and K. Nakani- 5 shi. 1999. A2E, a lipofuscin fluorophore, in human retinal We claim: pigmented epithelial cells in culture. Invest Ophthalmol 1. A method of treating retinal diseases, comprising: Vis Sci. 40:2988-2995. administering an effective amount of a composition com Toops, K. A., Tan, L. X., and Lakkaraju, A. (2014). A prising desipramine to a retinal disease patient, detailed three-step protocol for live imaging of intracel- 10 lular traffic in polarized primary porcine RPE monolayers. wherein at least one of a retinal disease symptom is Exp Eye Res 124C, 74-85. lessened or progression of the symptom is delayed and Toops, K. A., Tan, L. X. & Lakkaraju, A. Differential wherein the retinal disease is age-related macular regulation of organelle dynamics by tubulin acetylation in degeneration (AMD). 2. The method of claim 1, wherein the disease is associ polarized epithelia. J Cell Sci In preparation (2014). 15 Travis, G. H., Golczak, M., Moise,A. R., and Palczewski, K. ated with an abnormal accumulation of lipofuscin. (2007). Diseases caused by defects in the visual cycle: 3. The method of claim 1, wherein the composition is retinoids as potential therapeutic agents Amm Rev Phar applied topically to the affected eye or eyes. macol Toxicol 47, 469-512. 4. The method of claim 1, wherein the composition is applied in the following dose range: oral dose of between van Leeuwen, R., Tomany, S. C., Wang, J. J., Klein, R., 20 Mitchell, P., Hofman, A., Klein, B. E., Vingerling, J. R., 10-100 mg/day or topical doses of between 0.1 mg and 1 Cumming, R. G., and de Jong, P. T. (2004). Is medication mg/day. use associated with the incidence of early age-related 5. The method of claim 1, wherein the progression of the maculopathy? Pooled findings from 3 continents. Oph following symptoms is delayed: accumulation of soft, large drusen; fundus autofluorescence; and dark adaptation. thalmology 111, 1169-1175. 25 Wang, L., Cano, M., and Randa, J. T. (2014). p62 provides . 6. A method of treating macular degeneration in a subject dual cytoprotection against oxidative stress in the retinal m need thereof, comprising: pigment epithelium. Biochim Biophys Acta 1843, 1248- identifying a subject with macular degeneration; and 1258. administering to the subject a therapeutically effective amount of a composition comprising desipramine. Weng, J., Mata, N. L., Azarian, S. M., Tzekov, R. T., Birch, 30 D. G., and Travis, G. H. (1999). Insights into the function 7. The method of claim 6, wherein the macular degen of Rim protein in photoreceptors and etiology of Star eration is age-related macular degeneration. gardt' s disease from the phenotype in abcr knockout . 8. A method of treating macular degeneration in a subject mice. Cell 98, 13-23. m need thereof, comprising the steps of: identifying a subject with at least one of Xu, J., Toops, K. A., Diaz, F., Carvajal-Gonzalez, J. M., 35 Gravotta, D., Mazzoni, F., Schreiner, R., Rodriguez-Bou a) an accumulation of soft, large drusen, lan, E., andLakkaraju,A. (2012). Mechanism of polarized b) increased fundus autofluorescence, and lysosome exocytosis in epithelial cells. J Cell Sci 125, c) delayed dark adaptation; and 5937-5943. administering to the subject a therapeutically effective amount of a composition comprising an inhibitor of Yao, J., Jia, L., Shelby, S. J., Ganios, A. M., Feathers, K., 40 Thompson, D.A., andZacks, D. N. (2014). Circadian and acid sphingomyelinase activity, wherein the inhibitor is Non-Circadian Modulation of Autophagy in Photorecep desipramine. tors and Retinal Pigment Epithelium. Invest Ophthalmol 9. The method of claim 8, wherein the composition Vis Sci. additionally comprises a second inhibitor of acid sphingo- Zhao, C., Yasumura, D., Li, X., Matthes, M., Lloyd, M., 45 myelinase activity. Nielsen, G., Ahern, K., Snyder, M., Bok, D., Dunaief, J. 10. The method of claim 9, wherein the composition L., LaVail, M. M., and Vollrath, D. (2011). mTOR comprises at least one structural inhibitor of acid sphingo mediated dedifferentiation of the retinal pigment epithe myelinase activity and at least one functional inhibitor of lium initiates photoreceptor degeneration in mice. J Clin acid sphingomyelinase activity. Invest 121, 369-383. * * * * *